hckrnws

I'm happy for the guy, but am I jealous as well? Well yes, and that's perfectly human.

We have someone who vibe coded software with major security vulnerabilities. This is reported by many folks

We also have someone who vibecoded without reading any of the code. This is self admitted by this person.

We don't know how much of the github stars are bought. We don't know how many twitter followings/tweets are bought.

Then after a bunch of podcasts and interviews, this person gets hired by a big tech company. Would you hire someone who never read any if the code that they've developed? Well, this is what happened here.

In this timeline, I'm not sure I find anything inspiring here. It's telling me that I should rather focus on getting viral/lucky to get a shot at "success". Maybe I should network better to get "successful". I shouldn't be focusing on writing good code or good enough agents. I shouldn't write secure software, instead I should write softwares that can go viral instead. Are companies hiring for vitality or merit these days? What is even happening here?

So am I jealous, yes because this timeline makes no sense as a software engineer. But am I happy for the guy, yeah I also want to make lots of money someday.

It's funny to me how still so many don't realize you don't get hired for the best positions for being a 10x programmer who excels at hackerrank, you get hired for your proven ability to deliver useful products. Creativity, drive, vision, whatever. Code is a means to an end. If you're the type of programmer who thinks of yourself as just a programmer, and take pride in your secure code, ability to optimize functions and algorithms, you're exactly the kind of programmer AI will replace.

Quality of code has never had anything to do with which products are successful. I bet both youtube and facebook's codebase is a tangled mess.

> you get hired for your proven ability to (…)

No, you get hired for your perceived ability to (…)

The world is full of Juliuses, which is a big reason everything sucks.

https://ploum.net/2024-12-23-julius-en.html

Oh, Julius. Haven't we all met a Julius.

Story! Long ago, very long ago, I was working at a tiny Web company. Not very technical, though the designers were solid and the ops competent.

We once ended up hosting a site that came under a bit of national attention during an event that this site had news about. The link started circulating broadly, the URL mentioned on TV, and the site immediately buckled under the load.

The national visibility of the outage as well as the opportunity cost for the customer were pretty bad. Picture a bunch of devs, ops, sales and customer wrangling people, anxiously packed around the keyboard of the one terminal we managed to get logged into the server.

That, and Julius, the recently hired replacement CTO.

Julius, I still suspect, was selected by the previous CTO, who was not delighted about his circumstances, as something of a revenge. Early on, Julius scavenged the design docs I was trying to put together at the time to get the teams out of constant firefighting mode, and then started misquoting them, mispronouncing the technical terms. He did so confidently and engagingly. The salespeople liked him, at first.

The shine was starting to come off by the time that site went down. In a company that's too small for teams to pick up the slack from a Julius forever, that'll happen eventually.

So here we were, with one terminal precariously logged into the barely responding server, and a lot of national eyes on us. This was the early days of the Web. Something like Cloudflare would not exist for years.

So it fell on me. My idea was that we needed to replace the page at the widely circulated URL with a static version, and do so very, very fast. I figured that our Web servers were usually configured to serve index.html first if present, with dynamic rendering only occurring if not. So I ended up just using wget on localhost to save whatever was being dynamically generated as index.html, and let the server just serve that for the time being.

This was not perfect and the bits that required dynamic behavior were stuck frozen, but that was an acceptable trade-off. And the site instantly came back up, to the relief of everyone present.

A few weeks later, the sales folks, plus Julius, went to pitch our services to a new customer prospect. I bumped into one of them at the coffee machine right afterwards. His face said it all. It had not gone well.

Our eyes met.

And he said, with all the tiredness in the world: "He tried to sell them the 'wget optimizer'..."

This story made my day, thanks!

I've met countless Juliuses over the years. I kept track of the companies, and the Juliuses. My biggest revelation is that every company that was being in some substantial capacity led by a Julius (either at C level, VP, or high up in management) ended up one of two ways:

1. Shut down or shutting down (e.g. team reduced by > 50% since I've been there)

2. Julius removed, endlessly seeking work, keeps getting fired, and can't find a place to call home

The meteoric rise of the Julius is an exception - sooner or later their lucky streak ends and they face the cliff of adversity, towering above them with no way to climb it - no skills to help him actually do it.

I mean, maybe he was a revolutionary. One could describe what Vercel is selling as some kind of "wget optimizer" as well

In a couple of decades of work, I have never actually met anyone like Julius. Typically, I have found that those who excel at listening and presenting are also capable of understanding the technology at an appropriate level for their role -- it's not like this stuff is truly complicated, after all.

I have met quite a few people who are more focussed on the business than the technology, but those people tend to end up in jobs where the main problems aren't actually technical. Which, let's be honest, is the case in very many tech jobs.

oh man, I have met several Juliuses. one of them was my boss till he made an error as similar to the one the original Julius made, but unfortunately too late I had to leave the company earlier he made my life hell. now he is at another company, as long as he is at this company I won't apply there, if they hire him they have no place for me

No end of Juliuses. And they're not even the worst type you can meet at a software company.

There are so many. I think if you haven't met a Julius, chances are you are Julius..

I wonder, how does a Julius perceive another Julius, as another competent worker? What about a non-Julius then?

I have met armies of julius at all levels. Id say 80% of people are julius and if u dont think so then i have some news for you.

It is always like this. Your ability to socialize will bring you further than any other skillset. The Kennedys for example manufactured their status by socializing. Industry is no different.

Humans are social animals and good social skills is a major benefit almost everywhere, including at work. This does not make most people juliuses.

Hi julius! I kid and dont speak as if i am not a julius. Most people are in it for money for a house car family etc. they dont care about the job in so much as means to an end. That is julius but he took it further

> The Kennedys for example manufactured their status by socializing.

And generational wealth and serious political power.

https://en.wikipedia.org/wiki/John_F._Fitzgerald

https://en.wikipedia.org/wiki/Joseph_P._Kennedy_Sr.

Read some of the father’s quotes. He literally sent his kids out to marry into richer and more powerful families.

The question is how does one become julius

80% of people you meet are communicating to your customers that the server doesn't have an IP address for security reasons?

80% of the people are saying that this is highly complex software. We should not expect to serve more than 4 requests per second without a full kubernetes cluster backed by 27 pods, a cloud spanner database, and 200k lines of code.

I present, our contact form.

Julius is a metaphor for a specific type of person who is ignorant and useless but has mastered the way to appear otherwise.

If you think this was about IP addresses, well ...

And yet this thread is completely full of Frank Grimes.

That number feels off by a lot to me. I think i can say i'm quite good at socializing, quite above average when comparing to people I meet and work with. I'd rate my engineering skills about average level and i have a firm dislike of fraud and of people acting to be better/smarter/faster than they really are. In my career I've come across managers of the julius type, as well of the narcissistic type, even a sociopath. I would estimate 10 to 20 percent of people are of the Julius type.

It was a subtle ref to the 80/20 rule in that most people likely oscillate between the julius and the useful. Some of that 80% are full time julius for sure.

> Id say 80% of people are julius and if u dont think so then i have some news for you.

> Industry is no different.

Based on these comments, maybe some self-reflection is in order, as it seems from the 80% comment that what you mean is that 80% of people are able to adequately communicate.

Have i been the julius this while time? What a twist.

> perceived ability

In this case at least it's definitely more than that. Ever since LLMs became a thing, there has been a constant search to find it's "killer app". Given the steep rise in popularity, regardless of the problems, that is now OpenClaw. As they say, the proof's in the pudding; this guy has created something highly desirable by the many.

Yet, people are still asking for the usability of OpenClaw outside of marketing. It's a bit unclear how much of a "killer app" it really is, and how much is just burning money for the lulz and Bot RP. I personally also got the impression many people had their first AI-gateway experience with OpenClaw, and don't understand that those abilities have been around for a while now, but is located in the expensive LLMs which OpenClaw is using, not in OpenClaw itself. I've seen people thinking that OpenClaw is actually the AI.

> don't understand that those abilities have been around for a while now

Hugely underestimated comment. That's pretty much the entire point here. Many people didn't know something with these capabilities was already possible. Or some - like me - knew of the potential, but couldn't be bothered/didn't have the time to put the bits together in a satisfactory flow (I'm currently exploring and building on nanobot[0], which is directly inspired by OpenClaw; didn't touch OC because it's in JS and I'm a Python person). Everything came together really well, which is why it's a "killer app". And now the dam has burst there will be customized takes on the concept all over the place (I'm also aware of a Rust "port", Moltis[1]), taking the idea to next levels.

[0] https://github.com/HKUDS/nanobot [1] https://github.com/moltis-org/moltis

People weren't underestimating it and it's not that they "couldn't be bothered". They either understood the gaping security/safety holes it creates, or were guarded against their own stupidity.

Doesn't really matter. As always it's integration that makes a product.

Talking to bots on Telegram isn't new.

Running agentic loops isn't new.

Giving AI credentials and having it interface with APIs isn't new.

Triggering AI jobs from external event queues isn't new.

Parking state between AI jobs in temp files isn't new.

Putting all together in one product and marketing it to the right audience? New.

But novelty doesn't make a killer app. When outside of marketing and gateway-experience, there are still that many open questions, then maybe it's a valid claim to call it perception instead of substance.

At the end, only time will tell how much there really is to this.

>But novelty doesn't make a killer app.

It often does, if killer app means popular app.

There is a difference between popular and (in)famous. OpenClaw is famous, has popularity at the moment, but is it sustainable? Will OpenClaw (or some kind of successor) still have a relevant usage (outside of fan circles) next year? Or in 5 years?

And I'm not talking about just any kind of assistant, because those are already existing for decades now with various degrees of competence and all kind of flavours.

> Will OpenClaw (or some kind of successor) still have a relevant usage (outside of fan circles) next year?

I have a feeling OpenClaw et al. will only still exist if somehow all of the gaping security holes are ever able to be closed and through some sort of magic, less than 5% of the users get hacked within the next year, but I'm not sure it's even possible to close those holes, since the entire point and usefulness of such tools is to give them root access and set them completely free.

to--> latexr: Thank you for the link to Polum's essay in juliusosis. It really is the case that a lot of incompetence is hiding in plain sight. Probably because modern schooling encourages this.

I've lived in China (as a foreigner) and they have a word for Juliuses. They call them the 'cha bu duo xiansheng' = the 'Mr. Almost ok'.

I think you're right but you've been a bit pedantic about the parent comment. They sloppily said that delivering business value gets you hired, when in reality the appearance of that may do. But I think we all understood their main thrust was to disagree with the comment before them about coding ability, and the point is that this doesn't always correlate with business value.

I did enjoy your link though.

Your comment and the article expanded my world view a little bit. Thank you.

> Pour celleux qui ne connaissent pas l’informatique

https://shs.cairn.info/revue-cites-2020-2-page-137?lang=fr

This is if not the best article i have read recently. Julius ...

My imposter syndrome is essentially fear of being julius.

90% of software engineers have a fear of being Julius

How about none of the above, but hired because of wanting OpenClaw?

There's OpenClaw the codebase, and there's OpenClaw the community. They could build the same program very easily (as evidenced by the number of clones out there already). That part's not worth paying much for. But redirecting the whole enthusiast community around it? That's worth a lot.

Exactly. My point is that it might not be about the guy.

Wow, that blog post really gave me pause and has stuck in my head for the last hour or so.

Everything is perception though. You are looking at this with your own perception, biases, and heuristics just like everyone else. There is no 'right' way to hire.

Great article until the end when they talked about AI.

I haven't seen that before. But it was really hard get to the end. Not because it's bad written or so, on the contrary is a very good piece. However the feeling is unfathomable. I hate Julius'es. More so I hate the managers blinded by Julius'es.

Julius sounds like a sociopath. Sociopaths have no empathy/morals, so they can confidently lie all day and still be perfectly fulfilled; and some of them can be very excellent at social manipulation. This level of confidence in all things, including complete bullshitting, and constantly climbing the corporate ladder for huge payoffs, is not too uncommon among them.

IMO, all you can really do around one is try to focus on yourself. Or get away as fast as you can, depending on the situation.

Talk about going all the way to write the story and seeing the point go by

Your boss liked Julius. People liked Julius

You're not going to convince people they have to pay more attention to the technical guy that can't string a though together and answers in a grumpy mood

Be more like Julius and you might get more of his laurels

Nah. Avoid companies that can't see through the Juliuses. Because there will be other disastrous consequences to their bad decision making processes.

> Avoid companies that can't see through the Juliuses.

Good luck with that

I’m rather sure *Airbus* will prefer a programmer which reads and writes reliable code.

The programmer which delivers useful products is probably hired by Microsoft? Or worse, Boeing. Or Toyota. Some NTSB people or Michael Barr are happy to tell you details about the number of dead people they created.

    Restart braking to brake because our code failed.

    One single sensor delivers wrong data. Let us put the trim down. DOWN! DOWN!

    But I’m only programming a text viewer.

For Airbus, Boeing, and others the cost of failure is disproportionately high. Just look at how you consider Boeing despite that 99.99...% of their software and hardware work flawlessly. They will be known for the 737 Max failure for decades.

When OpenAI tells someone that suicide isn't that bad, some bs supplement could be the best thing to treat their cancer, or does anything else that has a negative outcome, the consequences are basically zero. That is even though any single failure like that probably kills alot more people per year than Boeing.

It seems there is knowledge of this and the lack of responsibility placed on these companies so they act accordingly.

But realistically, I just had 2 flights last month, checking what model of aircraft I was on didn't even cross my mind. I survived both flights btw.

There are only so many safety-first companies and products. The vast majority of the economy isn't optimizing for safety

> There are only so many safety-first companies and products

There are only so many companies that think of themselves as safety-first. In practice, basically all companies work on things that should be safety-first.

Does your software store user data? Congrats, you are now on the hook for GDPR and a bunch of similar data handling regulations.

Does your software include a messaging component? You are now responsible for moderating abusive actors in your chat.

Does your software allow users to upload images? Now you are a potential distribution vector for CSAM.

And so on... safety isn't just for things which can cause immediate death and dismemberment

There’s a difference between "safety matters" and “safety is the primary constraint". Most companies manage risk to an acceptable level while optimizing for speed and cost. Aerospace companies optimize for minimizing catastrophic failure, even at extreme expense. Treating a potential GDPR fine as equivalent to a flight-control failure ignores that society, regulators, and markets treat those risks very differently. The inconvenience and economic cost of your Discord messages leaking is not the same category of harm as your pacemaker controller failing. And because the majority of economic activity sits in that lower-criticality category, it would not be surprising if highly specialized, safety-critical human software engineering becomes more of a niche, while much of routine software development becomes increasingly automated or commoditized.

> Treating a potential GDPR fine as equivalent to a flight-control failure ignores that society, regulators, and markets treat those risks very differently

Agreed, though I think that if GDPR fines were actually being levied at the recommended 4% of global revenue, we'd start treating them more similarly to a 737 crash.

> The inconvenience and economic cost of your Discord messages leaking is not the same category of harm as your pacemaker controller failing

Sort of depends who they leak to. Your teen classmates who bully you to suicide? Your abusive ex who is trying to track you down to kill you? The 3-letter agency who is trying to rendition your family to an internment camp?

There are a lot of seemingly benign failure modes that become extremely lethal given the right circumstances. And because we acknowledge the potential lethality of something like a pacemaker failure, we have massive infrastructure dedicated to their mitigation (EMT teams, emergency external pacemakers, surgical teams who can rapidly place new leads, etc). For things society judges less important, mitigations are often few and far between

OT: it's not the first time I see this grammatical mistake: "didn’t trained". Is it some accepted regional variation?

I think he is a non-native speaker, like me. I also do this mistake very often and 'didn't train' is a bit counter-intuitive - at least for me.

Thank you! I assume “didn’t train” is correct. Probably my favorite mistake! I like it when people point out mistakes, give me corrections, and explain why. The reason is crucial.

Maybe “hadn’t trained” is even better. Makes sense when ordering times. But I don’t trust LLMs an inch. It makes up options for git[1] and both GCC and CLANG are often immediately telling me that the LLM is lying.

Cookieengineer and illichosky are right.

[1] Considering that man pages exist, it shows how useless their harmful crawlers are.

I think that happens when as a German you're used to using the Plusquamperfekt which is a somewhat unique tense that's allowed to be used in all past tenses.

It allows you to not having to define the point in time and neither the frame of the timespan's points in time.

Some languages allow to use that type of tense and it's somewhat a language gap I suppose. I have no idea what other languages or proto languages allow that tense though, but I've seen some Slavic and maybe Finnish(?) natives use that tense in English, too.

Maybe someone more elaborate in these matters has better examples?

[1] https://de.wikipedia.org/wiki/Plusquamperfekt

Airbus pays like shit probably. Just going off the stuff I've read about Boeing.

I can confirm, they do.

It's the interesting work tax, I guess. Though this doesn't look terrible: https://www.levels.fyi/de-de/companies/airbus/salaries/softw...

Ah, you were thinking of the German branch, I was talking about the French one, in Toulouse (I have a few friends working there).

There, a team lead is doing ~$4000 net per month. So not poverty, but not great either.

I'm not sure of the situation for software engineers but ones on the aerospace and mechanical side working in aerospace in Europe are paid something on the order of 50% less than ones in the US. I always assumed it's just a supply demand problem but I haven't run the numbers.

Did you factor in the highest cost of life (including housing, healthcare, retirement, etc.) in the US in your 50% ?

We're not talking about Airbus or centuries old commodified industrial companies. Airbus sells airplanes, not AI software tools.

But if you did build a core innovation in aerospace that went viral I'm sure Airbus would be interested in hiring you.

The salary would be 3K per month. And lunch coupons to buy a ham baguette.

I literally got my current cushy gig to fix a codebase that was crumbling under its own unmaintainable weight at a company that, like you, thought that quality doesn't matter. This is not the first time in my career I get a great job that way.

"Quality doesn't matter" people are why I'm not worried about employment. While there is value in getting features out fast, definitely, there always comes a point on your scaling journey where you have to evolve the stack structure for the purpose of getting those features out fast sustainably. That's where the quality of the engineering makes a difference.

(Anecdotally, the YouTube codebase may be locally messy, but its overall architecture is beautiful. You cannot have a system that uploads, processes, encodes, stores, and indexes massive amounts of videos every hour of every day that in the overwhelming majority of cases will be watched less than 10 times, and still make a profit, without some brilliant engineering coming in somewhere.)

Both can be true: people who deliver products based on vision and all are very much needed and cracked devs who excel in technical details as well. Peter and you are of these respective groups then.

The Youtube mobile app is a nightmare to use, and has been for months (desktop is working quite well but I am using my phone 95% of the time). Reopening a short shows me a few frames of the next video before freezing, shorts die on second play constantly, history crashes because of shorts, changing to videos brings them back but navigating to shorts crashes again.

This has been reliably going on for at least 6+ months, I thought shorts was a big priority for them, but the UX is and remains horrible.

This is where the debate has another axis - when.

Quality matters, delivery speed matters, shipping also matters, where it matters and when it matters is much harder to get right. But it's also self correcting - if you don't, the project or business die - you can only get it wrong for so much or for so long.

To only discuss on one axis is presumably why GNU Hurd have never shipped or how claude-c-compiler doesn't compile hello world.

You still need a few people high enough up in a company who think that quality does matter to be able to get the job to fix things.

That will happen, in the lucky cases, when someone high enough up with basic reasoning skills looks at support costs and time spent fixing bugs versus feature release velocity and sales income.

> If you're the type of programmer who thinks of yourself as just a programmer, and take pride in your secure code, ability to optimize functions and algorithms, you're exactly the kind of programmer AI will replace.

Hard disagree. I foresee the opposite being true. I think the ability to understand and write secure, well optimized, performant code will become more and more niche and highly desired in order to fix the mess the vibe coders are going to leave behind.

And the cheapest way to distribute that to everyone will be via AI coding.

If AI becomes good at doing that and fixing bugs, then sure. But there is no evidence pointing towards that as of now. Mostly only slop.

this is such a weird take to me. every piece of evidence I've seen shows that AI is quickly becoming better at writing code, debugging, finding security issues. my own experience, benchmarks, studies, new articles.. everything points to progress

Fully disagree.

There's lots of people that won't care about the code: executives, managers, customers etc. If the engineers don't care either, then who cares?

If we compare with big food companies, that's like their food formula. No one thinks it's useless - it's the source code for the product they sell. Yet nowadays we get so many engineers distancing themselves away from the code, like the software formula doesn't matter.

There are diminishing returns, but overall good code goes hand in hand with good products, it's just a different side of it.

Based on the interview format these days, I beg to differ.

If this were true, we wouldn't be studying Leet code and inverting binary trees to get a job.

I guess the lesson here is that unless you have a direct line with upper management to skip the line, you'll be stuck grinding algorithms for the rest of your life.

Leet code interviews are in the spirit of filtering out charlatans who misrepresent having even basic programming fundamentals. Many interviewers take it too far, but the original motivation is essential to saving time in the hiring process. I was instantly converted after participating in the full hiring process for a junior dev, which didn't properly filter for programming skill.

Big companies may have separate hiring SWE departments where the initial interviewers don't even know what team or role you may land in, so they have to resort to something...

I was nodding my head agreeing with you but then remembered John Carmack, who seems to deliver both... He takes great pride in writing ground breaking code, for industry defining products.

We should all try and be more like John Carmack.

The man is on a different level, cognitively speaking. That's like asking sprinters to "just be more like Usain Bolt". Some people are just built different. Carmack is one of them.

I admire the guy but he spends like 12 hours a day doing just that and his code is full of tricks, it's debatable as a paragon of quality. I don't think it's for everyone, to be Carmack, nor it should be; diversity is important.

I argue we shouldn't, because if everyone is like Carmack then no one is.

And only people on the older end of the spectrum have seen Carmack working in his element back in the day.

The things I want people to take from a guy like John Carmack, or Jon Blow, or Lukas Pope, or Ron Gilbert, or Tim Schafer, or Warren Spector, or Sam Lake, or David Cage god forbid...is pure curiosity and pushing the boundaries to make that real.

In every case there is a mix of a deep and unusual urge to make an idea happen with an affinity towards the technicality of it.

I bring Sam Lake into this because nobody has blended FMV with gameplay the way Remedy have and pushed the boundary on it.

Another detail is that his groundbreaking code was great part of made some of the products - I'm thinking of Doom.

It wasn't just for the sake of quality and best practices, it defined and had an impact on the product experience.

Like Doom probably wouldn't have been as successful if it was any other way.

The opposite is not true though: successful products might have messy codebases, but that doesn’t mean, that messy codebases lead to successful products, or that quality doesn’t matter.

There's a balance to strike, and it's hard to get right. You have to give up quality enough that you actually deliver things to users rather than working on 'the perfect code', but you also have to keep quality high enough that you're not slowed down by spaghetti code and tech debt so much that you can't deliver quickly as well.

This is made more complicated by the fact that where the balance lies depends on the people working on the code - some developers can cope with working in a much more of a mess than others. There is no objective 'right way' when you're starting out.

If you have weeks of runway left spending it refactoring code or writing tests is definitely a waste of time, but if you raise your next round or land some new paying customers you'll immediately feel that you made the wrong choices and sacrificed quality where you shouldn't have. This is just a fact of life that everyone has to live with.

I’ve met many more $5M/year “SaaS” entrepreneurs who built a Wordpress plugin than a custom SaaS platform. Your point is well made.

Right. Shopify apps, too, is a gold mine.

He's not hired to code. He has taste for "what works" in these types of things. They want him to apply that taste - maybe making new services or fixing old.

See: Rick Rubin.

"Rick Rubin says he barely plays any instruments and has no technical ability. He just knows what he likes and dislikes and is decisive about it."

https://www.cbsnews.com/news/rick-rubin-anderson-cooper-60-m...

https://en.wikipedia.org/wiki/Rick_Rubin_production_discogra...

> If you're the type of programmer who thinks of yourself as just a programmer, and take pride in your secure code, ability to optimize functions and algorithms, you're exactly the kind of programmer AI will replace.

I'm not sure how this follows logically from the comment you are replying to, which states:

> We have someone who vibe coded software with major security vulnerabilities.

I like your optimism but no - you are still hired via "excels at hackerrank", every big tech company first interview is exactly this, no matter how many projects your delivered and how useful you are/were at you previous job.

This seems to be largely an American phenomenon

In more minor markets like Europe/Australia it seems to be a lot less leetcode and a lot more (1) experience (2) degree (3) actual interview performance

AtlasSian? Canva? Absolutely the same process in Australia. Smaller shops/contractors - sure.

This is more so because the US companies have been flooded with East / South Asian workers. The proliferation roughly tracks with a decrease in white (European) American representation in tech companies. US companies used to be much more like you described.

> Quality of code has never had anything to do with which products are successful. I bet both youtube and facebook's codebase is a tangled mess.

The code’s value is measured in its usefulness to control and extend the Facebook system. Without the system, the code is worthless. On the flip side, the system’s value is also tied to its ability to change… which is easier to do if the code is well organized, verified, and testable.

> If [you] ... take pride in your secure code

I don't object to most of what you're saying, but I take issue with this part.

This happens to be an area where lapse or neglect can be taken as a moral failure. And here you are mocking people who are concerned about it.

If someone uses AI to architect a bridge and the bridge collapses, you couldn't say that the structural integrity of the bridge wasn't the important part.

Would you feel comfortable flying on an airplane where the programmers don’t care about secure code, correctness, or the ability to reason about and optimize algorithms—where “good enough” is the philosophy? Most people intuitively say no, because in safety-critical and large-scale systems, engineering rigor isn’t optional. Software may look intangible, but when it runs aircraft, banking systems, or global platforms, the same discipline applies.

The “Facebook/YouTube codebases are a mess so code quality doesn’t matter” line is also misleading. Those companies absolutely hire—and pay very well—engineers who obsess over security, performance, and algorithmic efficiency, because at that scale engineering quality directly translates to uptime, cost, and trust.

Yes, the visible product layers move fast and can look messy. But underneath are extremely disciplined infrastructure, security, and reliability teams. You don’t run global systems on vibe-coded foundations. People who genuinely believe correctness and efficiency don’t matter wouldn’t last long in the parts of those organizations that actually keep the lights on.

Do you think the people writing the code that operates aircraft care about code quality? After the boeing incident I do not.

Fair point and that’s exactly why Airbus has been eating Boeing’s lunch. When engineering culture takes a back seat to cost, schedule, and optics, outcomes diverge fast. In safety-critical systems, rigor isn’t optional, it’s the competitive advantage.

I find it difficult to believe software is Airbus’ competitive edge. First, their software for aircrew bidding is an absolute and utter disaster. Date filtering has been broken nearly a year despite multiple releases being pushed. Date management is like THE KEY functionality of aircrew bidding. I also use their flight plan software and it’s like they never bothered to ask a pilot how they use a flight plan in flight.

I think Airbus is riding the coat tails of solid engineering done in the 80s and continuing to iterate that platform vs Boeing trying to iterate on a hardware platform from the 60s. Airbus benefited significantly from 20s years of engineering and technological progress. Since the original design of the A320, changes have been incremental. Slightly different engines, addition of GPS/GNS, CPDLC, CRT to LCD screens. Meanwhile Boeing has attempted to take a steam gauge design from the 60s and retrofit decades of technology improvements and, critically, they attempted to add engines significantly altering the aerodynamics of the aircraft.

> Quality of code has never had anything to do with which products are successful.

It may look like that, but many of the products with bad code didn't even make it into your vibe statistics because they weren't around for long enough.

This is exactly right.

The goal is delivering a useful product to someone, which just requires secure enough, optimized enough, efficient enough code.

Some see the security, optimization, or efficiency of the code itself as the goal. They'll be replaced.

As long as AI can't make the code optimized and secure by itself, and these day it still can't, those people won't be replaced. And when they do get replaced there is no guarantee that the more "entrepreneur" population won't get replaced as well.

Except it wasn't and still isn't secure enough.

Yes, Facebook's early PHP code looks pretty bad by today's standards

Facebook PHP Source Code from August 2007: https://gist.github.com/nikcub/3833406#file-index-php

> you get hired for your proven ability to deliver useful products.

Ah, right. Write "Brew", which gets used by thousands of devs at Google every day, and then get rejected in an interview.

> your proven ability to deliver useful products

Which is not the case. It's just a useless product, without any real use case, which also introduces large security bugs in your system.

> you get hired for your proven ability to deliver useful products

Huh, if you make finished products you better start your own company.

And yet most companies don’t hire primarily for vision and creativity. They need far more people who can execute someone else’s vision reliably. You can’t neither win the battle nor the war with only generals.

Visionaries are important, but they’re a small part of what makes a successful organization. The majority hinges on disciplined engineers who understand the plan, work within the architecture, and ship what’s needed

As Victor Wooten once said: "If you’re in the rhythm section, your job is to make other people sound better." That’s what most engineering positions actually are and there’s real skill and value in doing that well.

> If you're the type of programmer who thinks of yourself as just a programmer, and take pride in your secure code, ability to optimize functions and algorithms, you're exactly the kind of programmer AI will replace.

The most successful engineers are the ones who can accurately assess the trade-offs regarding those things. The things you list still may be critical for many applications and worth obsessing over.

The question becomes can we still achieve the same trade-offs without writing code by hand in those cases.

That’s an open question.

>It's funny to me how still so many don't realize you don't get hired for the best positions for being a 10x programmer who excels at hackerrank, you get hired for your proven ability to deliver useful products

For a programmer, that's based on them "being a 10x programmer who excels at hackerrank".

For manager types it might be "Creativity, drive, vision, whatever".

>Code is a means to an end

For a business in general.

When hiring developers, code IS the end.

But it also looks like these companies value and pay for the tech bro version of a snake oil consultant. And that you still have to have a lot of things going in your favour for your own brand of slop to elevate you to tech celebrity status. I don't see anybody who isn't already well-connected or financially comfortable pulling this off because nobody who has something to lose will slop their way to the top.

I don't think it's a good thing that the craft of software engineering is so easily devalued this way. We can quite demonstrably show that AI is not even close to replacing people in this respect.

Am I speaking out of envy or jealousy? Maybe. But I find it disappointing that we have yet more perverse incentives to hyper-accelerate delivery and externalise the consequences on to the users. It's a very unserious place to be.

> Code is a means to an end.

Product is a means to an end.

Being good at something is a means to an end.

That end? Barter for food and shelter, medicine.

The means to do so; code or delivery of a product; are eventually all depreciated, and thrown away. You eventually age into uselessness and die.

Suddenly having an epiphany it's not about code but product! way too late in the game, HN... you're just trying to look like you got it figured out and bring deep fucking value to humanity right as "idea to product without intermediary code layer" is about to ship[1]. You already missed your window.

You still don't get the change that's needed and happening due to automation; few of us want to put you on their shoulders and sing songs about you all.

Hop off the Hedonistic Treadmill and get some help.

[1] am working on idea to binary at day job, which will flood the market with options and drown yours out

>you get hired for your proven ability to deliver useful products

Tell that to the guy that made brew and tried to interview at Google

Delivering a product is one thing. Continuing to upgrade it and maintain it indefinitely is another. Good quality code makes it easier to make improvements and changes as time goes on. Doesn’t matter if you’re a human or an LLM.

Also, has anybody looked through the Openclaw source? Maybe it’s not so bad

> Quality of code has never had anything to do with which products are successful. I bet both youtube and facebook's codebase is a tangled mess.

This is such a bad take and flat out wrong. Your ability to deliver and maintain features is directly impacted by the quality of the code. You can ship a new slop project every day if you like, but in order for it to scale or manage real traffic and usage you need to have a good foundation. This is such a bad approach to Software engineering.

> you get hired for your proven ability to deliver useful products

Or, in this case, just because they need a poster boy for their product, which isn't as good as they say it is.

It took me a while to realise that most people don't care how it's done or how it works they just want something useful and working (even if it's vibe coded or duct taped)

You also believe that AI will replace mathematicians?

This is so not true.

You are replying to someone whose account name is tabs_or_spaces, which in itself is so ironic that I have no word for it.

What people don't seem to realize is that like you pointed out there's a demand for the previous "developer relations" type of job though, and that job kind of evolved through LLM agents into something like an influencer(?) type position.

If I would take a look at influencers and what they're able to build, it's not that hardcore optimized and secured and tested program codebase, they don't have the time to acquire and hone those skills. They are the types who build little programs and little solutions for everyday use cases that other people "get inspired with".

You could argue that this is something like a teacher role, and something like the remaining social component of the human to human interface that isn't automated yet. Well, at least not until the first generation of humans grew up with robotic nannies. Then it's a different, lower threshold of acceptance.

Should I be sad or rather relieved that grifters will be able to grift without my help? I would just accept the reality and reeducate myself to some other field where hard engineering is still required, but I'm afraid AI will advance faster than my degree.

> Quality of code has never had anything to do with which products are successful

This is just wrong. Plenty of examples of crap code causing major economic losses.

Exactly, quality of code is one of those necessary but not sufficient things... If you are somehow successful without quality of code (e.g. early Twitter maxing Rails performance) you end up either crash and burning of spending crazy amounts on infrastructure/rewrites (and often both).

>>It's funny to me how still so many don't realize you don't get hired for the best positions for being a 10x programmer who excels at hackerrank

Competitive coding is oversold in this generation. You can log in to most of these sites and you will see thousands of solutions submitted to each problem. There is little incentive to reward situations where you solved some problem which a thousand other people have solved.

To that end its also a intellectual equivalent of video game addiction. There is some kind of illusion that you are indulging in a extremely valuable and productivity enterprise, but if you observe carefully nothing much productive actually gets done.

Only a while back excessive chess obsession had similar problems. People spending whole days doing things which make them feel special and intelligent, but to any observer at a distance its fairly obvious they are wasting time and getting nothing done.

Tell that to the creator of Homebrew, Max Howell

> "Google: 90% of our engineers use the software you wrote (Homebrew), but you can’t invert a binary tree on a whiteboard so fuck off."

Yeah you’re right, the engagement factories probably don’t care about code quality. The customer isn’t the customer after all.

I mean, you're right but at the same time you're talking about something completely different. Software with security vulnerabilities is not a useful product. You don't address the raised issues.

...huh?

10x programmers aren't the ones grinding hacker-rank.

Neither are the programmers like me who actually focus on building good systems under any significant threat.

And Facebook's codebase is pretty decent for the most part, you'd probably be shocked. Benefits of moving fast and breaking things include making developer experience a priority. That's why they made Hacklang to get off PHP and why they made React and helped make Prettier

I think you are really just describing an outlier. Most people really do get hired for the first thing. This is a situation where someone went viral and got a job. I don't think this is sort of the rule. The thing about "proven ability to deliver ..." is just kind of cope recruiters tell themselves and other people. It's nice but its not how things cache out in the real world.

> It's telling me that I should rather focus on getting viral/lucky to get a shot at "success".

A vibe coder being hired by the provider of the vibe coding tools feels like marketing to sell the idea that we should all try this because we could be the next lucky ones.

IMHO, it'd be more legitimate if a company that could sustain itself without frequent cash injections hired them because they found value in their vibe skills.

Someone that makes vibe coding tools would presumably want to have vibe coders on staff? If you're just not into the whole enterprise that's one thing but I'm not understanding what's fishy about that.

pets.com moment

Nah, I'm getting more of a webvan vibe...

The irony about Webvan, it was a good idea, but too early.

Kinda like the Apple Newton

Online grocery delivery was successful in the UK in the 1990s — Tesco started online ordering in the same year (1996) as Webvan, but could use their existing supermarkets as warehouses so avoided one of Webvan's main problems.

My parents used it occasionally, and I remember them/us demonstrating it to other parents. The software was supplied on a CD-ROM, and it connected to the internet only to download the stock list and place the order.

Peter has been prolific and talented long before AI tools. I became familiar with his work a decade ago: https://github.com/steipete/PSTCollectionView

People seem to think that because we all have the same tools and because they’re increasingly agentic, that the person wielding the tool has become less relevant, or that the code itself has become less relevant.

That is just not the case, at least yet, and Peter is applying a decade plus of entrepreneurial and engineering experience.

He was recently interviewed on Pragmatic Engineer, a podcast whose guests almost always have very impressive technical careers (the episode before him was Mai-Lan Tomsen Bukovec, the VP of Data and Analytics at AWS and the episode after him is Brady Gooch, the Chief Scientist for Software Engineering at IBM)

I agree that summarizing Peter as a "vibe coder" is unfair and disingenuous. The podcast paints his career as being interesting because we went from an impressive software developer, to an entrepreneur, to taking a significant break, to kind of obsessively creating Clawdbot.

Worth a listen https://newsletter.pragmaticengineer.com/p/the-creator-of-cl...

You are most likely confusing OpenClaw with Moltbook, which is the project that had the most glaring vulnerabilities. But even if OpenClaw was full of holes it would not matter.

Peter is not just a random "vibe coder" and he does not need to be hired by OpenAI to achieve "success". Before this he founded and sold a company that raised €100M. It is not his first project in the space either (see VibeTunnel for instance).

OpenAI is not hiring him for his code quality. They are hiring him because he proved consistently that he had a vision in the space.

What vision? Everyone and their mother has been trying to build useful AI assistants and personal CRMs since computers were invented - way before LLMs. He glued it together, and he succeeded because he executed before anyone else.

I applaud what he's done, and wish him luck trying to get this working safely at scale, but the idea that he's some visionary that has seen something the rest of the world hasn't is ludicrous.

Not Moltbook, ClawHub. Over 15% of ClawHub skills were malicious at one point, including the most downloaded. And they haven't even tried to solve prompt injections.

ClawHub isn't even useful. You can just point tell your OpenClaw agent you want it to do, and it will implement it. No need to rely on someone else's code^H^H^H^H textual descriptions of how to do talk to service xzy.

OpenAI, Anthropic, and other model providers have created tools (the LLMs) with unprecedented new capabilities. The key problems are a) these new tools have weird limitations that make them hard to deploy effectively, and b) these tools are so fundamentally new that creating useful products out of them is an exercise in discovery and requires incredibly novel, forward-thinking vision.

Pete, more than anyone in the OSS community IMO, exemplifies both of these qualities. He is living very much on the bleeding edge, so yes, the 10s of projects he's shipped faster than most devs can ship 1 are not as polished as if he'd created them by hand. But he's been pushing the envelope in ways that few, if any, are, and I'd argue that OpenClaw is much more the result of Pete living on that edge and understanding the trade-offs of these tools better than just about anyone.

Personally, I'm much more jealous of the fact that Pete has already had a successful exit under his belt and had the freedom to explore & learn these tools to the fullest. There is definitely a degree of luck involved with the degree to which OpenClaw took off, but that Pete discovered it is 100% earned IMO.

He also spent 13 years building [an] OCR document engine company (PSPDFKit) before becoming an "overnight" vibe coder success story.

His PDF toolkit was pretty solid and high quality if you were in the iOS space.

He’s not just a “vibe coder”.

There's an excellent Changelog podcast interviewing him which talks about his early career as well.

You don't get hired for any of that. OpenAI did not hire him because he went viral. Virality brought something interesting to OpenAI's attention. And they thought they could use this product idea/vision/execution, GTM strategy, whatever, and because it didn't seem like OAI had anyone on their team capable of this, they hired him.

Simple as that. Don't feel jealous, trying to replicate won't work, he did not know he'd be hired, he built something that he found interesting, and then realized it would be interesting to a lot more people.

The way to reach success is to either be strategically consistent in a way that maximizes luck surface area but does not depend on it, or to be unexpectedly lucky. The latter is gambling, People win the lottery regularly, does not mean you should make that your mission.

Be comfortable with not being the one to hit gold. And yes, it's ok to be jealous. Take a moment and then go back and enjoy the rest of your life.

Finally, there are a lot of companies that would likely hire you, hoping to hit gold. But you are likely filtering them out because they're not tech/large/startupy enough for you. These companies are wondering what they need to attract talent like you.

This should be a wake up call. A product's value is not a function of its code elegance. Nobody who matters notices the code, or cares. This is hugely inspiring to the most lazy+clever engineers, because it frees up so many thinking calories. Instead of trying to perfectly choreograph every bit of architecture to optimize for 1M concurrent users, you can spend 0.1 of the time and get things out the door, where you learn if spending even a minute of your time was worth it. Even better, when you realize tech debt is something that never needs to be paid down, you can focus all your energy on evolving your thinking patterns, not being bogged down in refactoring things that you've spiritually moved on from. An engineer's time is so precious; it needs to be spent thinking, not coding.

The guy has a long history of building popular products, long before vibe coding became possible. He is certainly good at writing code manually as well.

I genuinely think people on HN are having the misconception that vibe coding == don’t care about (the quality of) the code.

I like to think it’s the same as delegating implementation to someone else.

Except there are literally people on this thread saying that this is proof that code quality doesn't matter and that we all need to "wake up". It's the same as the people saying that spec driven development is the way of the future and that engineers should be focusing on the spec and not even looking at the code.

If you use LLMs and you do care about code quality, then great. But remember that the term vibe coding as it was coined originally meant blindly accepting coding agent suggestions without even reviewing the diffs.

Many of the people aggressively pushing AI use in code are doing so because they care more about delivery products quickly then they do about the software's performance, security, and long-term maintainability. This is why many of us are pushing back against the technology.

[dead]

I see a guy who has shown evidence that he has the skill and agency to successfully ship and scale a project that people want, pushing the frontier tools to their limits. That is valuable.

> a project that people want,

do many people actually use openclaw (a two week old project IIUC) or is it just hyped up?

Judging by the Openrouter-Rankings it is No. 1 or 2 in token usage, depending on the view (daily, weekly, monthly).

Most LLM usage does not go through Openrouter. Most people access LLMs thorough ChatGPT, Gemini etc. apps, integrations into popular products…

The percentage of total tokens being handled by Openrouter is a tiny blip

Many people use it.

Also he made a few other products, some of which were used probably by more than a billion people.

I bet they did not invert a binary tree on the whiteboard, nor answered how many golf balls fit into a plane.

Really surprised by all the comments here, they didnt hire him because of the amazing security openclawd had, but because he's one of the first one's who made a truly personal assistant that's actually valueable to people.

It's about what he created, not what he didnt create.

They're not acquiring the product he built, they're acquiring the product vision.

Also surprised; building something people want and proving it is the unlock. HN first principle since the beginning.

If you read his blog you’ll find about a lot of his engineering decisions.

Peter was right about a lot of the nuances of coding agents and ways to build software over the last 9 months before it was obvious.

Was he? Openclaw is now dead, right? The software will now die. No-one's going to maintain it.

This was a short-term gain for a long term loss.

I remember in the web 3 era some team put together a CV in one page site, literally a site that you could put your linkedin, phone no and email on but pretty, bought for millions.

Was the product a success or the marketing? As the product was dead within weeks.

There's a lot of low hanging fruit in AI at the moment, you'll see a few more things like this happen.

> No-one's going to maintain it.

Why? He's going to maintain it and the community is large enough. Another sci-fi idea that's slowly becoming real is that the project is maintaining itself.

OpenClaw is a bunch of projects that evolved together (vibetunnel, pi-mono, all the CLIs). It's even more interesting to see the next iterations, not only what happens to this project.

We've been here a 1,000 times. This exact news happens over and over again on HN. Why would it be any different from the other times this happens?

This is what US tech companies do to stay dominant. Buy and kill.

This project will die now, that's the point of buying him. For super cheap too but the sound of it.

Openclaw is an open source project. that's the beauty of the Open source. the community can take over and people can fork it. there already many clone of openclaw.

do you know about www.linktr.ee ? :-D :-D

It's not about the code, it's about the vibe.

Also, Peter is quite well known in the dev circles, and especially in mobile development communities for his work on PSPDFKit. It is not like he's some unknown developer that just blew up - he owned a dev tooling company for over 10+ years, contributed a lot to the community and is a great dev.

Pete didn’t just vibe code, he took his many years of engineering experience and applied it to build a ton of products, pushing the boundaries of todays models and harnesses.

I am saddened that the top post is about jealousy, do so many people feel this way? Jealousy should be something that when we feel we reflect on privately and work on because it is an emotion that leads to people writing criticism like tbis that is biased due to their emotional state.

If you just commit AI generated code without even looking at it it doesn't matter how many years of engineering experience you have.

You focused only on the past few months of his career, but this is just the tip of the iceberg. He was active for more than a decade, from early iOS development days to having a fairly successful exit.

So after almost two decades of hard work, it is not really fair to say he just vibe-coded his way into OpenAI.

Peter clarified “I don’t read code” part in Lex Fridman interview - he said “I don’t read the boring parts of the code” that are about data transformation or writing to/reading from databases.

He distinguished between what he calls “Agentic Engineering” and “Vibe coding”, and claimed majority of the time he is not just Vibe coding.

He has 80,000+ GitHub contributions in a year across 50+ projects. I’m not sure how he averages 200 commits per day by just looking at diffs from a terminal, but it’s just Superhuman - https://github.com/steipete

> Then after a bunch of podcasts and interviews, this person gets hired by a big tech company. Would you hire someone who never read any if the code that they've developed? Well, this is what happened here.

I have a feeling that OpenAI and Anthropic both use AI to code a lot more than we think, we definitely know and hear about it at Anthropic, I havent heard it a lot at OpenAI, but it would not surprise me. I think you 100% can "vibe code" correctly. I would argue, with the hours you save coding by hand, and debugging code, etc you should 100% read the code the AI generates. It takes little effort to have the model rewrite it to be easier to read for humans. The whole "we will rewrite it later" mentality that never comes to pass is actually possible with AI, but its one prompt away.

Boris has been very open about the 100% AI code writing rate and my own experience matches. If you have a typescript or common codebase, once you set your processes up correctly (you have tests / verification, you have a CLAUDE or AGENTS.md that you always add learnings to, you add skill files as you find repeatable tasks, you have automated code review), its not hard to achieve this.

Then the human touch points become coming up with what to build, reviewing the eng plans of the AI, and increasingly light code review of the underlying code, focusing on overall architectural decisions and only occasionally intervening to clean things up (again with AI)

I'm aware I've been building my own alternative to Beads for weeks now. the instructions.md file and something like Beads is epic.

https://github.com/Giancarlos/guardrails

I didnt like how married to git hooks beads was, so I made my own thats primarily a SQLite workhorse. Been using it just the same as I have used Beads, works just as good, drastically less code. I added a concept called "gates" to stop the AI model from just closing tasks without any testing or validation (human or otherwise) because it was another pain point for me with Beads.

I fully sync the issues to GitHub to boot.

https://github.com/Giancarlos/guardrails/issues

Works both ways, to GitHub, from GitHub. When you claim a task, its supposed to update on GitHub too (though looking at the last one I claimed, doesnt seem to be 100% foolproof yet).

I think you’re conflating things. You probably are not jealous but rather frustrated and coming from a point of a false dichotomy trying to equal your position to his. If you were to stop and actually compare your lives you would likely find very different humans. It’s easy to fall into this trap sometimes, don’t let it get into your head. Be grateful for being you and enjoy what life has to offer you instead.

If you want to make a million bucks a year then go put in three consecutive quarters of demonstrable lift on a renenue-adjacent metric at Stripe or Uber.

If you want to make a zillion a year ask Claude to search for whatever Zuckerberg is blowing a billion on this quarter.

All of those companies are certain to exist in 12 months. Altman is flying to Dubai like every other week trying to close a hundred billion dollar gap by July with a 3rd place product and a gutted, demoralized senior staff.

Going by how insanely viral OpenClaw has been on X, I don’t think any of the stars were bought

There were some comments somewhere below about that virality being bought though. I don't know how true that is or where those commenters got their information. If you look at google trends though there is practically no mention of ClawdBot before around January 23, even though the project was released in November.

It was renamed many times. It was also called "clawdis" at one point, and prior to that "warelay," when it was simply a Whatsapp gateway for Claude Code. It was already gaining some momentum at that point but wouldn't reflect as search results for "Clawdbot," and especially wouldn't be visible on Google Trends when most of the conversation was on X/Github.

Even if the conversation was on other sites, people would still search for it.

he likely poured oil into the flame investing a few hundred bucks to double the virality

and according to HN, somehow that's a bad thing? Like making your own project more popular and successful using your own resources is...bad?

Fake engagement doesn't need to be bought anymore.

This person created a bot factory. It's safe to assume that most of the engagement is coming from his own creation. This includes tweets, GitHub stars, issues and PRs, and everything else. He made a social network for bots, FFS.

He contributed to the dead internet more than any single person ever. And is being celebrated for it. Wild times.

> He made a social network for bots, FFS.

Matt Schlicht made Moltbook, not Peter Steinberger.

Well, you got me. That changes everything.

> Then after a bunch of podcasts and interviews, this person gets hired by a big tech company

I think the whole OpenClaw arc has been fun to follow, but this sudden turn away from OpenClaw and toward the author as a new micro-celebrity that ended with OpenClaw being sidelined to a foundation was not what I saw coming.

Congrats to Pete for getting such an amazing job out of this, but it does feel strange that only a few days ago he was doing the podcast circuit and telling interviewers he has no interest in joining AI labs.

I don’t think this story arc should be seen as something replicable. Many have been trying to do the same thing lately: Hyping their software across social media and even podcasts while trying to turn it into cash. Steve Yegge is the example that comes to mind with his desperate attempts to scare developers into using his Gas Town (telling devs “dude you’re going to get fired” if they don’t start using his orchestration thing). The best he got out of it was a $300K crypto pump and dump scam and a rapidly dropping reputation as a result.

Individuals who start popular movements have always been targets for hiring at energetic companies. In the past the situation has been reversed, though: Remember when the homebrew creator was rejected from Google because he didn’t pass the coding interview? (Note he later acquiesced to say that Google made the right call at the time). That time, the internet was outraged that he was not hired, even though that would have likely meant the end of homebrew.

I do think we’ll be seeing a lot of copycat attempts and associated spam promoting them (here on HN, too, sadly) much like how when people see someone get success on YouTube or TikTok you see thousands of copycats pop up that go nowhere. The people who try to copycat their way into this type of success are going to discover that it’s not as easy as it looks.

Well, once you learn that hard work does not pay, it’s really your own fault if you keep believing in it.

What matters is the result, not how hard you worked at it. Schools and universities have been teaching this for a long time, that what matters is just a grade, the result.

Read his backstory. He’s a high quality software engineer by background.

> vibe coded software with major security vulnerabilities

> vibecoded without reading any of the code

Remember when years ago people said using AI for critical tasks is not an issue because there is always a human in the loop? Turns out this was all a lie. We have become slaves to the machine.

I’m surprised to read this comment. I totally get why openAI hired the guy, IMO its a brilliant hire and I wish Meta would have fought more to get him (at the same time Meta is very good at copying and I think they need more people pushing products and experiments and less processes, they’ve been traumatized by cambridge analytica and can’t experiment anymore)

I feel similar... OpenClaw has lots of vulnerabilities, and it's very messy, but it also brought self-hosted cron-based agentic workflows to your favorite messaging channel (iMessage, telegram, slack, WhatsApp, etc.), which shouldn't be overlooked

They're buying him for his ideas, not for his ability to code. And if his stars are bought, then they're buying him also for his black hat marketing, I guess...

He didn't even have to be the one buying them. Lots of people benefit from a tool like OpenClaw getting popular.

He didn't specify the role he was hired for, code is just a means to an end. Perhaps OpeaAI wanted him for his vision (I like to think so) or just to make up for the public support they're losing (I hope not). In either case, it may not be an engineering role.

The bit about purchased stars and followers is a bit out of left field. Is there a piece of news I missed?

Good for him! But it is possible he won't stay there for a long time. Like Geohot at Apple. There is a difference between working on a fun project which you completely control and being under a constant pressure and having to follow constrains and requirements set by managers in a corporation.

It's telling me that I should rather focus on getting viral/lucky to get a shot at "success". Maybe I should network better to get "successful". I shouldn't be focusing on writing good code or good enough agents.

All of this is true and none of it is new. If your primary goal is to make lots of money then yes you should do exactly that. If you want to be a craftsman then you'll have to accept a more modest fortune and stop looking at the relative handful of growth hacker exits.

I was jealous too until I realized this is just an ad for OpenAI. They want to show you can vibe code an app and actually become a millionaire. What better way to show than actually do it?

Well, here you have it, a low effort to wire up a few tools together with spaghetti gen ai and he’s millionaire in a few months. Ok, I might be mean by saying no effort, I actually don’t know. But I know vibe coding won’t work for more than a few weeks. Also I think this bot is just a connector to multiple open source libraries that connect to WhatsApp and other services.

This is the best ad to sell AI: you can be millionaire too if you use our ChatGPT to vibe code stuff.

I think it will get a negative reaction in a few weeks when the dust settles as technical people realize it’s an ad.

Note: he might be an amazing developer but the ad still stands.

Edit: from Gemini: Publicly Embarrassing Anthropic: The timing is brutal. Anthropic’s legal team forcing a name change (from "Clawdbot" to "Moltbot" to "OpenClaw") alienated the very developer who was driving millions of users to their model. OpenAI swooping in to hire him days later frames Anthropic as "corporate lawyers" and OpenAI as "friends of the builders." It’s a perfect narrative victory.

Jealousy is exactly the reaction they are hoping to trigger: use our tools, build something popular, get paid out. What better marketing spend than buying this project.

You could always get, mh, lucky. That is the most common startup exit plan: Finding someone who pays for half a business or an idea. Now it happens more quickly. Everything does.

But that path was never about writing good code.

I don't know this guy's abilities so can't comment on that, but looking at how much AI companies spend on marketing - that's a great hire.

It does not matter that he vibe-coded it. It does not matter if any stars/twitter post were bought. He generated hype and that's what big AI company need at the moment. They hire him, they give a cut on that hype. If he's no good (at generating any hype) in the coming months, he'll be gone. It's hype all the way down.

This isn't a surprise at all. I sat down with the dev team at OpenAI during dev day last year and the biggest shocker to me: these "kids" are over here vibe coding the whole damn thing.

See that as winning the "startup lottery", that doesn't mean what he did is rational or smart, he just had a great outcome.

In trading it's the same, you can make stupid bets and make a lot of money, doesn't mean you're good trader.

Nothing to conclude from this, this kind of hype-fueled outcome has always been a part of life.

I was half-jokingly telling someone the other day (before I knew what OpenClaw was or anything about this story), that as the ability to code is becoming commoditized, sales and marketing skills are going to be more important, shifting power from techies to influencers and we may see Mr Beast become a software powerhouse.

> It's telling me that I should rather focus on getting viral/lucky

This is the real dangers of social media and other platforms. I know teachers in the school system, way too many kids want to grow up to be influencers and YouTubers, and try to act like them too.

At the risk of sounding like an old man yelling at the sky, this is not good for society. Key resources and infrastructure in our society is not built on viral code or YouTubers, but slow click of engineering and economic development. What happens when everyone is desperately seeking attention to become viral? And I don’t blame the kids the influencers by nature show a very exciting or lavish lifestyle.

    > way too many kids want to grow up to be ... YouTubers

    > way too many kids want to grow up to be influencers

> To me, if you look closely at (social media) influencers, they are nothing more than people who were popular in high school and managed to extend it for a few years with the use of social media.

That's a very superficial similarity. It's one thing for a kid wishing to be popular in their extended social circle, and a very different thing a young adult being convinced that they can "grind" their way to influencer fame and money.

The young adult may never have heard of or considered the extreme survivorship bias in the stories of successful influencers.

> What's wrong with wanting to be a YouTuber?

Which YouTubers are we talking about here? Hobbiests? People chasing social clout? People who like making stuff and sharing it? People pushing negativ social attitudes? Context matters.

I’m talking about young adults not preparing for their future because they think they are going to become millionaires on YouTube, they focus on what is essentially a culture of grifting, with a success rate similar to winning a lottery.

Im not sure what has to change, but the current state of things is not healthy.

You don’t need the lucky shot. But luck needs room to happen. What you need to grow into is becoming a leader. Mentor others, lead by example, suggest new things and build prototypes for show and tell. All that is actually the growth path for good senior software engineers, not becoming a middle manager creating Excel sheets.

And that’s more or less all he did. Had an idea, build a prototype, showed to the world and talked about it - even inspired people who are now saying „I could have done that“. Well do it, but don’t just copy. Improve the idea and great something better. And then very early share it. You might get lucky.

What he built is genuinely interesting even if it is not something I would want to give all my credentials to. Makes sense for OpenAI to hire someone who has shown he can build something a lot of people want even they don’t know how to make an even half secure app out of it. They probably think he has the right judgement of where UX would need to move to. That is easily more valuable for them than any coding.

> vibecoded without reading any of the code

Isn't this the actual definition of vibe coding?

In my view this is just an aquihire to get a headline and take ownership over this trend. Yet another pivot to build hype.

> We have someone who vibe coded software with major security vulnerabilities. This is reported by many folks

> We also have someone who vibecoded without reading any of the code. This is self admitted by this person.

Peter was pretty open about all of this. He doesn't hide the fact. It was a personal hack that took off and went viral.

> We don't know how much of the github stars are bought. We don't know how many twitter followings/tweets are bought.

My guess, from his unwillingness to take the free pile of cash from the bags.fm grift, is that this in unlikely. I don't know that I would've been able to make the same decision.

> Then after a bunch of podcasts and interviews, this person gets hired by a big tech company. Would you hire someone who never read any if the code that they've developed? Well, this is what happened here.

Yes, I'd hire him. He's imaginative and productive and ships and documents things. I can fix the code auditing problem.

> In this timeline, I'm not sure I find anything inspiring here.

Okay?

> It's telling me that I should rather focus on getting viral/lucky to get a shot at "success".

Peter has been in the trenches for years and years, shipped and sold. He's written and released many useful tools over the years. Again, this was a project of personal love that went viral. This is not an "overnight success" situation.

> So am I jealous, yes because this timeline makes no sense as a software engineer. But am I happy for the guy, yeah I also want to make lots of money someday.

Write and release many, many useful tools. Form a community and share what you're building and your chances will greatly increase?

I mean some might say that's like joining a sinking ship. Of course one man's trash is another man's treasure. To each their own.

Hiring in tech has been broken for many many years at this point. There's so much noise and only more noise coming now with AI. To be completely honest it's entire random from my end when hiring. We can't review every application that comes in. It's just impossible. We do weed out some of the spam of course and do get to real people that actually fit the requirements, but there's so many other talented people who would easily fit the role that simple get buried under applications. It's depressing from all sides here. No one should think that they aren't any good or did something wrong or didn't network enough... because the unfortunate truth is that getting a job in tech is a lottery. Something many don't want to admit.

Im working on a project to change this.

Funny that you mention 'real people'. There are a number of components that sit at the core of what Im building - it should allow you to have the time and reach to vet more (100% verified) candidates than you ever could before. I also want to reduce the cost of hiring so that firms can hire more people.

>We have someone who vibe coded software with major security vulnerabilities. This is reported by many folks

>We also have someone who vibecoded without reading any of the code. This is self admitted by this person.

And we have a company whose product should adhere to the highest security standards possible, hiring this guy.

I've seen the same result play out a few times on LinkedIn - random person studying for an MS in CS or AI, blogs and posts about stuff they're vibe coding with Lovable or whatever, builds a decent following, and then, from tagging various AI-related firms, lands a job at one of them.

The field has kind of been like this for a while - people with portfolios of proven work done, showcasing yourself and your personality via blogs or vlogs makes you sort of a known quantity, versus someone with just a CV and a LinkedIn page.

This is yet another example of an area where extroverts have an advantage. You could be 10x the engineer that the creator of OpenClaw is, but that's irrelevant in this timeline if nobody has ever heard of you.

Errr its always been extremely true that social networking brings success. With far more value return than writing great code nobody knows about or uses.

Would you like to have it removed? https://github.com/oswarld/openshears

I don't think he is hired for coding, he is hired for the product. It is not that he is going to join a product team and code, he probably will lead and influence the product, where other software engineers can help to fulfill.

Vibe coding is just a tool - same with programming languages and compilers.

The product being useful and well received by user and market is still the ultimate test. Whether something is vibe coded or not does not matter.

Whenever technically more capable folks diss the growth of a non technical person into bigger roles, I'm obligated to post this Steve Jobs video being asked about Java.

https://www.youtube.com/watch?v=oeqPrUmVz-o&t=6

As I understand Peter had already early retired because of a successful startup exit and presumably has more money than he knows what to do with. Does that help make you feel less jealous on him getting a job at oai?

It's the old story: evil, irresponsible behaviour has a higher chance of success, than being good and responsible. AIs recent history is a good example. Google had the lead, but lost it (temporary) to OpenAI, because Google was responsible and were not willing to open pandoras box. Apple seems to have something similar to OpenClaw for a while now, but withholds from releasing it, because it's too unsecure. History is full of people burning the world for their own greed, and getting rewarded for it; and they then call it "taking risks" and "thinking outside the box"... I think the underlying reason might be in too many people thinking there is some level of competence behind the irresponsible behaviour and it's alls just controlled harm or something like that.

I think it is unfortunate that this is happening. After all the mishaps and wrongdoings I don't want see anyone joining openai

If all the above is true, why didn't Sam & Co just replicate his product and offer their own improved version - - with security incorporated within ?

Because he wanted to say "Take that" To anthropic who forced him to change the name of the product.

> We don't know how much of the github stars are bought. We don't know how many twitter followings/tweets are bought.

Why this insinuation? The project went massively viral and was even covered in my local newspaper. I don't see any reason to doubt those numbers.

As if newspapers never did paid promotion articles?

Yeah, the small local newspaper for my town of 30k is a paid shill for OpenClaw. Amazing comment.

I wouldn't necessarily expect him to be hired as their lead developer, but I think he would be a good product manager. He's clearly created something people want and see potential for.

> It's telling me that I should rather focus on getting viral/lucky to get a shot at "success"

Kids and young people have known this forever at this point. Sadly.

Maybe think of this as a hiring of a marketer and tech influencer. And someone with the chops to create a viral product.

Exactly.

If AI companies believe code generate by it self, people to scaling up sales is the only worth hire.

> It's telling me that I should rather focus on getting viral/lucky to get a shot at "success".

I'm pretty sure that's meant to be the general lesson of the last 20 years or so in Silicon Valley, but it's just survivorship bias in action.

You don't hear a whole lot about the quietly successful engineers who work a 9-till-5 and then go home to see their wife and kids. But you do constantly hear about the folks who made it big YOLO'ing their career and finances on the latest a startup/memecoin/vibecoded app...

Exactly. This whole thing just seems like a repeat of Flappy Bird to me. What was the "lesson" of Flappy Bird for game developers? That you should make very small, very simple games? How has that worked out for the vast majority of copycats who tried? The truth is there isn't any lesson, other than "sometimes people play the lottery, get lucky and win". Most people who play won't, though.

No need to be jealous. If you'd have watched some of the interviews of this guy then you'd know that he's not vibe coding.

Such is life in the attention economy.

https://www.youtube.com/watch?v=_95AKKmqGvE

Semantics and grammar joke aside.. there are not many workers remembered in history. Only the so-called absolute greatest, meanest, etc are remembered. Nobody remembers the people who worked on the pyramid, but everyone knows some Farao.

In this case they hired someone who has 'mastered' the use of their own tool(s). Like if Home Depot hired a guy who has almost perfect knowledge of each and every tool in their own portfolio.

I'm not really sure if i want to be that guy.

Ugh. Have we all forgotten that jealousy is the absolute opposite of a good virtue. Why does this get upvoted? Hacker News in a truly despicable state these days when this is what bubbles to the top. It saddens me to see that all the good people here have left or stopped participating. When we hear how rotten social media is, this also includes HN.

dont be jealous. working for some evil corporate is soulsucking for most humans. Only few thrive in such environments. most will try to get quick $$ and exit before they feel completely dead inside.

not sure why i find a lot of these types of comments lately, just a sign of the times i guess? criticism sure, but to reduce all of his work as if it were a paragraph prompt or something, that's something else.

i hate when the people start bringing up the "luck" factor as if you are the only smart one here to realize that it also plays a huge factor?

you want to make lots of money? change your mindset, stop making excuses and roll the dice. it won't guarantee success, but i also guarantee nobody who did so would ever lament how unfair it was that they worked so hard and someone else succeeded through "luck" so they might as well not try.

Do good or at least useful work in public and you'd be surprised at what can happen.

It’s not the code. It’s the vision and the can-do attitude. And perhaps a bit of the (earned) name.

nah. focus on building cool things people want.

I'm more jealous of his muscles and butt

> It's telling me that I should rather focus on getting viral/lucky to get a shot at "success".

Well duh. I thought that was well understood.

The other option is having well-to-do parents a la Musk or Gates.

Have you tried that?

> It's telling me that I should rather focus on getting viral/lucky to get a shot at "success".

It doesn't? You'd need to know the odds for the tell. Like how many incompetent grifters are there, how many of them become hugely successful?

> Would you hire someone who never read any if the code that they've developed?

I mean, if I'm a company specifically in the business of selling to companies the idea that they can produce code without reading any of it? Yeah, obviously I'd hire them.

> We also have someone who vibecoded without reading any of the code. This is self admitted by this person.

This is isn’t right. He says very clearly in the recent Lex Fridman podcast that he looks at critical code (ex: database code). He said he doesn’t look at things like Tailwind code.

Don't forget he also had Sam Altman's phone number. Do you any of you have his number? Also before he did all this he was semi retired for 5 years because of a successful exit. So for anyone thinking they can replicate this ask...

1. Are you already rich? Do you have cash in the bank to vibecode a project fulltime for many months just for fun?

2. Do you have Sam Altman's (or similar) number?

He didn't create or release something as finished.

He built something and shared it.

People took liberties with it.

It's not about getting viral/lucky... it's about enjoying experimenting and learning.

Money follows your unique impact and imprint in these kinds of cases.

Maybe it still is supposed to sound fancy to say you didn't read any of the code. The guy definitely could very deeply understand, read and edit the code, he developed the industrial standard liberal for PDF editing (used by Dropbox etc).

Just saying what you want might be the future for development of some kinds of software, but this use case sure seems like a very bad idea.

I very much appreciate the vision he put into practice, but feel sorry for the project being acquihired kind of.

The lesson here is to make something people want. All else is forgiven is the product is something people really want - the product market fit most of us never achieve.

That wasn‘t true before.

https://news.ycombinator.com/item?id=15713801

This is such a strange take to be the top comment of “hacker” news. Why are we shaming someone who “hacked” something together and made it open source?

It's also from a guy who rebranded three times (Clawdbot, Moltbot, OpenClaw) in a row and this is technically his fourth rebrand.

Props for admitting jealousy and for being honest! I often feel the same way when fixing bugs in others code.

What sense it made to do something like Instagram? There were already N social networks where you could share photos. No technical excellence was needed. It was just momentum, being in the right incubator, and so forth... I understand what you are saying, but it has been always like that.

Well - no. There are some products where the product itself was relatively simple to build, and the rest was product-market fit. Those are the easy ones technically, but that's not the only type of successful product. YouTube wouldn't be working today if it broke all the time under load.

it's a tough pill to swallow for developers, but nobody cares about your ability to write code. people care about you shipping something people want.

i can easily hire 100 sweatshop coders to finetune your code once i have a product that works but the inverse will never happen

That's such a bizarre thing to claim when offshoring software development has historically been a huge failure. You've always needed competent technical staff with even more demanding management requirements to stand a chance.

What percentage of programming job interviews every went like that? They ask fizz buzz, they ask DP, they system analysis and design, and some culture fit. Maybe some people might ask this B-school type stuff but who is out there verifying deliverables of people from previous jobs?

Well you don’t see the real value of coding tasks during interview. What gets tested are your communication skills, how you think and express your thinkings. You will be working in a team so you need at least fit and work with others. You are right that no one cares about your FizzBuzz.

One day Atlas may shrug, but not today, atlast..

HN Really hates understanding business. All these comments, yet no one has gotten the answer right.

OpenAI bought marketing and now someone else cannot buy openclaw and lock out Openai revenue from a project that is gaining momentum.

There are a many of these business moves that seem like nonsense.

1. Bought for marketing.

2. Adversarial hire. ie hire highly skilled people before your competitors can even if you don't have anything for them do to. Yet...

3. Acqu-hire. Buy a company when you really just want some of the staff.

4. Buy Customers. You don't care about the product and intend to migrate their customers to your system.

5. Buy competition before its a threat.

[dead]

Well that was a crazy month. Kudos to this guy for recognising his goals which is not to start another company. It is very easy to get intoxicated by the idea of something being so successful that you can capture the value, especially after having struggled for so long with a previous company. I think it's every founder's dream to like just hit lightning. But this stuff is incredibly stressful and it's important to be able to look into the future and ask yourself. Is this what I want? Is this what I need in my life? And the answer here is no. This person can deliver value elsewhere quite easily and get the reward without as much stress. We should all take a lesson from this whirlwind journey. Do not attempt to be like Peter. You can admire the work he's done. Do not attempt to replicate it. Appreciate it for what it is. For yourself as an observer or a user it's a lesson. But also to note that this is an anomaly. You will never replicate it. A lot of people feel a little bit of envy or jealousy. I used to feel that when I was working on something and I saw other people succeed and I wished that that had happened to me. But if it was meant for you it would find you. And the fact that it hasn't found you means that it was not meant for you. We all have our role to play. There is something important for us to do and that's not necessarily something that is world famous or amasses thousands of GitHub stars. If after reading this it's still bothering you. Take a walk and reflect on the good things in your life.

This feels like such a defeatist take. The ideas time had come. For luck to strike you have to be in the market for it. Just keep shipping and playing. We don’t “all have our role to play” but there are a lot of roles that need playing

But that's the point it's not defeatist. It's more about saying there is something for you to do. There is a role for you to play but that's not necessarily the role that you see somebody else playing. So sometimes we see somebody else doing something and we think whoa that person is successful and we become envious of that and then we want to emulate that. But we forget that maybe that's not what's intended for us. Maybe that person is really good at that thing or that's what was for them. But for us we might be good at something else and there might be something that we are uniquely positioned to do so. The point is not to be defeatist but not to focus on what somebody else has right. Focus on what you have and focus on what your ability is and focus on what's going to improve your quality of life and the people around you and don't focus on the negative aspects of what is effectively fomo

Appreciate your take. I think we’re bouncing around the same mental state from different sides.

I do not believe in predetermined roles. My version is to find the thing you’re excited to do and not the outcome you’re excited to have.

Yea sorry it probably comes across as pre-determined but you as a person have likely spent a long time becoming good at something and have a certain personality and experience based on your life, so I guess what I'm saying is, that sort of creates a role for you, and when you understand what that is, you can really hone in and do good work. Sometimes its not obvious to us and when we see something we might want to go after it. That's fine. I guess my point is, don't look at the shiny thing and chase that. It's not what's going to fulfil you in life e.g Peter probably wasn't chasing the shiny thing. He was trying to solve a personal problem and it resonated with a lot of people. But when people see something be successful or this kind of wildride where you end up with a hugely successful project and go to a huge company like OpenAI, they focus on the wrong things. The inner insecurity takes over and you wonder, why not me, and how do I do that. But essentially it comes back to, solve problems. Solve interesting problems, work on things that you think are meaningful, and whatever the success might be, that's for someone else to decide. But I think people chase "fame", cause that's what we essentially see. Validation through popularity. It won't fulfil you. Trust me. But yes to your point. We're coming at the same thing from different angles.

What if the thing that excites you can't pay you money, and you find your overall life unsatisfying because of the things you have to do to earn money? Sorry, went a bit off rails there.

> Do not attempt to be like Peter. You can admire the work he's done. Do not attempt to replicate it. Appreciate it for what it is. For yourself as an observer or a user it's a lesson. But also to note that this is an anomaly. You will never replicate it.

This can be said about a lot of successful projects, products, and companies. I’d argue that, by all means, do try to be like Peter. Try to tinker around and make something new the world has never seen before.

He made something that excited many people, and I don’t think it’s the correct take to consider this an anomaly. It’s someone who was already known in the development community trying something new and succeeding.

Good move. OpenClaw is alpha quality, very dangerous, super useful and super fun - which amplifies the danger. It’s a disaster waiting to happen and a massive risk for a solo dev to take on. So best to trade it for a killer job offer and transfer all that risk.

To get a sense of what this guy was going through listen to the first 30 mins of Lex’s recent interview with him. The cybersquatting and token/crypto bullshit he had to deal with is impressive.

He's not really trading anything though. He was hired by OpenAI. OpenClaw will remain free and open source (it's the first line of his blog post). He says that OpenAI will allow him to work on it and already sponsors it so maybe that means he'll have time to improve it, I guess.

Given he's moving to SF to work in their office I presume part of it is he'll be working in-house on their commercial replacement, and will continue to cover costs on the OSS version which he's free to work on. His recent posts make it clear they've got plans for their own stuff to replace it.

Right, he indicated he's losing $10-20k a month on OpenClaw. Might as well let OpenAI absorb those losses and join the rocket ship.

The guy is a multi millionaire from selling his old software so I'm doubtful it is about the money for him at this point as much as it is the experience working with this tech on another level.

Yes, just like Jony Ive has been doing

Lex's interview https://youtu.be/YFjfBk8HI5o.

It really is quite funny though isn't it. Yes, it's a fucking hand grenade that will blow up at any moment. It's perfect for a one man band startup. Because there's massive upside and at the end of the day if it blows up he's just back to sqaure one, what did you expect from a 1 man band.

But wait. Here he comes. Hero of the hour. Sam Altman.

Let's take that wildly dangerous, lightly thought through product, and give it the backing of the leading AI lab. Let's take all that pending liability and slap it straight onto the largest private company in AI.

OpenClaw is more like an art project than a consumer product. It has shown clear consumer product demand. The next step is making it a safer consumer product.

> OpenClaw is more like an art project than a consumer product

I think this is true for a lot of vibe coded applications. Never thought about calling it an art project but it hits home.

You definitely have a point there.

> Listen to... Lex

People can do that? I always assumed Lex was a CIA psyop to experiment with the ability to make people sleep on demand.

While that might take it a little too far, Lex surely is a dangerous individual. On various occasions did he sympathize with the war and terror that Russia is doing in Ukraine. I do not click on any of his content because I will not support these (and a few other questionable, to say the least) views of his. Also his image of an MIT researcher is hilarious.

Yes. How do so many people fall for this guy? I find him pretty creepy, to be honest.

I'm a big fan of his. I particularly enjoy his long technical interviews like the one he did with Peter.

He's just another alt right podcast grifter.

This is the same "heating" effect as social media algorithms apply to random podcasters (e.g. Joe Rogan) - those isolated cases of success which happen to be completely synthetic provide an 'american dream' for the system, whose success depends on the Fantasy being alive and believed in by those who are its customers/product

I like the parallel as Joe Rogan is a trained actor who mastered the art of incorporating all the success factors of its predecessors. He saw obscure podcasts gaining intense viewership, he literally mimicked the patterns and merged it into the "best" of all. Even made his more mainstream, while fooling millions to feel they are part of a niche enlighten resistance community.

I recall listing to one of the now vintage series, I thought it was Joe Rogan himself. But it wasn't, the voice was a bit different but the pause, the reactions, the "waaah" with the overall tone of uncovering some secret truth.

It's a fascinating societal phenomenon, coupled with the American dream, yes.

In any case those examples are doing no good by setting themselves as models for millions to become obsessed in replicating. No surprise the rate of people in depression keeps going up.

To Rogan's credit, he was early to the concept of video podcasts, which Youtube's algorithm picked up on after they lifted video time limits in 2010 (or thereabouts). His videos were so long that he also profited handsomely from pre-roll and mid-roll ads that Youtube started introducing early in the last decade. Finally his popularity exploded once he started delving into "pop culture commentary" and "conspiracy theories" in the mid-2010s.

If you step back and look at this whole thing from a marketing and cash flow perspective, I think it makes a lot more sense.

It is in OAI's best interests to create a perception that flinging agentic swarm crap at the wall may result in lucrative job offers. Or to otherwise imply this is the golden path. They need their customers to consume ever more tokens per unit time. This highly contentious parallel agent swarm stuff is the perfect recipe.

Very good business observation.

Plus employees who can inject hyped ideas is exactly the sort of efficient advertising openai relies on

It will hurt when self proclaimed coders realise 2 years later and all their savings burned on token they cannot all get meaningful traction.

With OpenClaw we are seeing how the app layer becomes as important as the model layer.

You can switch models multiple times (online/proprietary, open weight, local), but you have one UI : OpenClaw.

> You can switch models multiple times (online/proprietary, open weight, local), but you have one UI : OpenClaw.

It’s only been a couple months. I guarantee people will be switching apps as others become the new hot thing.

We saw the same claims when Cursor was popular. Same claims when Claude Code was the current topic. Users are changing their app layer all the time and trying new things.

Memory. I have built up so many scripts and crons and integrated little programs and memories with open claw it would be difficult to migrate to some other system.

System of record and all.

Considering you have built them all in last few weeks, it should not be that difficult and no reason other systems won't reuse same.

Exactly! The whole point of personal agents is that the data is yours and it's where you want it not in someone's cloud. What harness you use to work with this should be a matter of preference and not one of lock in.

The future will be ownership of our memories and data. AI companies will fight tooth and nail to keep that data walled in and impossible to export.

Since it's already not walled-in in most cases I don't see this happening very effectively.

Using openrouter+kilocode I can simply switch between different providers' models and not miss out on anything.

If regulators force the capability of exporting to exist, what ya gonna do?

I continue to find it amusing that people really think corporates are really holding power. No - they are holding power granted to them by the government of the state.

Remind me why Zuck et al had to kiss the ring.

Very often, the regulators don't. Here in the US, half the country would refinance their mortgage for iMessage interoperability... if it were possible. Any time regulators reach for the "stop monopoly" button, Tim Cook screeches like a rhesus monkey and drops a press release about how many terrorists Apple stops.

If lobbying was illegal then you might have a point here, but alas.

There will definitely be migration tools.

The new agents might have a feature to query your old agents for a migration.

That said, I find it really hard to believe that you've generated so much work in the past few weeks since OpenClaw launched that you could never migrate to something else. It hasn't been that long.

How hard do you think it would be for ai to generate all those for some alternative?

AI didn't do the work, I did. Building up context is the part we actually have to put work into. I'm not saying it would be impossible, but boy would it be annoying to have to constantly reach a new assistant about your whole life.

"Here's my corpus of records from OpenClaw. Please parse it and organize into your own memories" boom done

Why are you assuming you'd have to do the work yourself?

This is a perfect use case for a new agent to query the old agent and get the details.

You could have OpenClaw summarize and export them into a format that the new one wants.

Maybe the new agents will be designed to be compatible with OpenClaw's style.

There is no reason to believe that you're locked in to something.

Bring your system to my records.

The irony of systems of record is that if there is more than one, there are effectively none. Just data stuck in silos waiting for compute.

Have you heard about this thing called AI coding agent....

Unless I am mistaken, that is all plain old markdown, arguably the easiest to migrate format for such data there can possible be.

Heck, that was half the pitch behind Obsidian, even if the project someday ended, markdown would remain. And switching between Obsidian and e.g. Logseq shows the ease of doing so.

Sorry but for $5 in credits you can have an agent port over all your bullshit to the next fad. I'll have one port over all my bullshit when the time comes too.

This effect isn't that important while the customer base is growing fast.

You can use OpenClaw to migrate these scripts off OpenClaw.

[dead]

Indeed, coding agents took off because of a lot of ongoing trial and error on how to build the harness as much as model quality.

This is the sort of thing employers are failing on. They sign contracts that assume employees are going to be logging in and asking questions directly.

But if I don’t have a url for my IDE (or whatever) to call, it isn’t useful.

So I use Ollama. It’s less helpful, but ensure confidentiality and compliance.

It’s only 2 months and there are already a rush of viable alternatives, from smaller, lightweight versions, to hosted, managed SaaS alternatives.

I’d suspect the moat here will be just as fragile as every other layer

openclaw is just one of many now, there are new ones weekly.

And OpenClaw is nothing revolutionary. It’s all shit we could do before OpenClaw. It’s just that no one was stupid enough to do it. Now everyone has gone crazy.

Plus you can get the model to write you a bespoke one that suits your needs.

I've been digging into how Heartbeat works in Openclaw to bring directly into Vibetunnel, another of Peter's projects

AI is the new Javascript?

Why?

You can literally ask codex to build a slim version for you overnight.

I love OpenClaw, but I really don't think there is anything that can't be cloned.

There’s actually many UI’s now? See moltis, rowboat, and various others that are popping up daily

I think the point was about the frequency of switching your frontend. With a proper frontend you can switch the backend on each request if you want, but usually people will stay with one main-interface of their choice. For AI, OpenClaw, Moltic, Rowboat are now such a frontend, but not many will use them all at once.

It's similar to how people usually only use one preferred browser, editor, shell, OS.

Are there any with a credible approach to security, privacy and prompt injections?

Does any credible approach to prompt injection even exist?

Anyone who figures out a reliable solution would probably never have to work again.

Not that I'm aware of, but I probably won't be interested in these kinds of assistants until there are.

This has been the case since the beginning of last year imo

Seems like models become commoditized?

Same for OpenClaw, it will be commodity soon if you don't think it is already

It appears to me that the same people who think “vibe coding” is a great idea, are the same people who think “Gas Town” is the future, and “OpenClaw” detractors are just falling behind.

For your sake, I’m not saying they’re wrong. I’m just pointing out something I’ve noticed.

It's definitely not right now. What else has the feature list and docs even resembling it?

OpenClaw has only been in the news for a few weeks. Why would you assume it’s going to be the only game in town?

Early adopters are some of the least sticky users. As soon as something new arrives with claims of better features, better security, or better architecture then the next new thing will become the popular topic.

OpenClaw has mediocre docs, from my perspective on some average over many years using 100s of open source projects.

I think Anthropic's docs are better. Best to keep sampling from the buffet than to pick a main course yet, imo.

There's also a ton of real experiences being conveyed on social that never make it to docs. I've gotten as much value and insights from those as any documentation site.

Not sure. I mean the tech yes definitely.

But the community not.

The community is tiny by any measure (beyond the niche), market penetration is still very very early

Anthropic's community, I assume, is much bigger. How hard it is for them to offer something close enough for their users?

> Anthropic's community, I assume, is much bigger. How hard it is for them to offer something close enough for their users?

Not gonna lie, that’s exactly the potential scenario I am personally excited for. Not due to any particular love for Anthropic, but because I expect this type of a tight competition to be very good for trying a lot of fresh new things and the subsequent discovery process of new ideas and what works.

My main gripe is that it feels more like land grabbing than discovery

Stories like this reinforce my bias

It has already been so with ppq.ai (pay per query dot AI)

I mean, ppq.ai (which I’ve never heard of) had zero to do with the commoditisation of LLMs. The industry did that. And services like OpenRouter are far more serious and responsible in this area than this ppq.ai is.

Things that arn't happening any time soon but need to for actual product success built on top:

1. Stable models

2. Stable pre- and post- context management.

As long as they keep mothballing old models and their interderminant-indeterminancy changes, whatever you try to build on them today will be rugpulled tomorrow.

This is all before even enshittification can happen.

This is the underrated risk that nobody talks about enough. We've already seen it play out with the Codex deprecation, the GPT-4 behavior drift saga, and every time Anthropic bumps a model version.

The practical workaround most teams land on is treating the model as a swappable component behind a thick abstraction layer. Pin to a specific model version, run evals on every new release, and only upgrade when your test suite passes. But that's expensive engineering overhead that shouldn't be necessary.

What's missing is something like semantic versioning for model behavior. If a provider could guarantee "this model will produce outputs within X similarity threshold of the previous version for your use case," you could actually build with confidence. Instead we get "we improved the model" and your carefully tuned prompts break in ways you discover from user complaints three days later.

? We saw this years/months ago with Claude Code and Cursor.

But it just codes. And are console / ide tools.

Openclaw is so so so much more.

That’s missing the point. OpenClaw is just one of many apps in its class. It, too, will fall out of favor as the next big thing arrives.

What’s the moat exactly?

None. That's why joining openai is the perfect fit.

Well, duh.

You being able to go places is the interesting thing, your car having wheels is just a subservient prerequisite.

This is all about PR now.

openclaw is inevitable type of software (as cli agents, as context-management software, as new methodologies of structuring sofware for easier AI ingestion, etc). Guy gamed, built it, guy got it.

At this point I would not expect well-rounded software as a byproduct of huge investments and PR stunts. There will be something else after LLMs, I bet people are already working on it. But current state of affairs of LLMs and all the fuss aroud them is way more peceptive, PR and emotion driven than containing intristic value.

Congrats to Peter!

Can any OpenClaw power users explain what value the software has provided to them over using Claude code with MCP?

I really don’t understand the value of an agent running 24/7, like is it out there working and earning a wage? Whats the real value here outside of buzzwords like an ai personal assistant that can do everything?

It has a heartbeat operation and you can message it via messaging apps.

Instead of going to your computer and launching claude code to have it do something, or setting up cron jobs to do things, you can message it from your phone whenever you have an idea and it can set some stuff up in the background or setup a scheduled report on its own, etc.

So it's not that it has to be running and generating tokens 24/7, it's just idling 24/7 any time you want to ping it.

Like what exactly? Can you give example of the type of prompts you are sending for the agent to do?

The messaging part isn’t particularly interesting. I can already access my local LLMs running on my Mac mini from anywhere.

AI companies must hate this right? Because they're selling tokens at a loss?

Google has started banning account that use Antigravy's discounted access instead of paying full price for the API https://github.com/openclaw/openclaw/issues/14203

> Impact: > Users are losing access to their Google accounts permanently > No clear path to account restoration > Affects both personal and work accounts

honestly, this is why I would not trust gemini for anything. I have a lot tied to my gmail, I'm not going to risk that for some random ai that insists on being tied to the same account.

Using different Google accounts won't save you, once Google decide to ban for TOS, all related accounts go with it https://news.ycombinator.com/item?id=30823910

They blocked your entire Gmail/Google account , not just the Gemini access?

That's a recipe for bots to ruin a lot of people's life.

Bold of you to assume profitability is one of their KPIs

My understanding was that if everyone paid and used AI the companies would go into liquidation on energy bills etc

Energy bills wouldn't be the problem if everyone used AI, energy supply would be.

Are you sure? I thought tokens (or watts) were sold at such a loss that if current supply limits were reached they’d go broke

The entire marginal cost to serve AI models is paid for by the API costs of all providers by nearly every estimation. The cost not currently recouped is entirely in the training and net-new infrastructure that they're building.

These companies are generally profitable for inference but it does not cover the cost of R&D (training).

From all indications the big players have healthy margins on inference.

Research and training are the cost sinks.

Is that just because people pay subscriptions and never use their tokens? Same model as ISPs

As an experiment, I set it up with a z.ai $3/month subscription and told it to do a tedious technical task. I said to stay busy and that I expect no more than 30 minutes of inactivity, ever.

The task is to decompile Wave Race 64 and integrate with libultraship and eventually produce a runnable native port of the game. (Same approach as the Zelda OoT port Ship of Harkinian).

It set up a timer ever 30 minutes to check in on itself and see if it gave up. It reviews progress every 4 hours and revisits prioritization. I hadn't checked on it in days and when I looked today it was still going, a few functions at a time.

It set up those times itself and creates new ones as needed.

It's not any one particular thing that is novel, but it's just more independent because of all the little bits.

So, you don't know if it has produced anything valuable yet?

It's the same story with these people running 12 parallel agents that automatically implement issues managed in Linear by an AI product team that has conducted automated market and user research.

Instead of making things, people are making things that appear busy making things. And as you point out, "but to what end?" is a really important question, often unanswered.

"It's the future, you're going to be left behind", is a common cry. The trouble is, I'm not sure I've seen anything compelling come back from that direction yet, so I'm not sure I've really been left behind at all. I'm quite happy standing where I am.

And the moment I do see something compelling come from that direction, I'll be sure to catch up, using the energy I haven't spent beating down the brush. In the meantime, I'll keep an eye on the other directions too.

> Instead of making things, people are making things that appear busy making things.

Sounds like a regular office job.

Yeah I'm not sure I understand what the goal here is. Ship of Harkinian is a rewrite not just a decompilation. As a human reverse engineer I've gotten a lot of false positives.This seems like one of those areas where hallucinations could be really insidious and hard to identify, especially for a non-expert. I've found MCP to be helpful with a lot of drudgery, but I think you would have to review the llm output, do extensive debugging/dynamic analysis, triage all potential false positives, before attempting to embark on a rewrite based on decompiled assembly... I think OoT took a team of experts collectively thousands of person-hours to fully document, it seems a bit too hopeful to want that and a rewrite just from being pushy to an agent...

That sound like being a manager IRL.

$3 z.ai subscription? Sounds like it already burned $3k

I find those toys in perfect alignment with what LLM provider thrive for. Widespread token consumption explosion to demonstrate investors: see, we told you we were right to invest, let's open other giga factories.

How's it burned $3k on a $3/month subscription running for a few days?

I simply don't get how it could have run for quite a while and only cost $3. Z.ai offers some of the best model out there. Several dollars per million tokens, this sort of bot to generate code would burn millions in less than 30 minutes.

They have a coding plan

And the $3 plan also has significant latency compared with their higher tier plans.

Keep us posted, this sounds great!

What a great use of humanity's adn the earth's resources.

Not being tied to Anthropic’s models and ecosystems, having more control over the agent, interacting with it from you messaging app of choice.

There's some neat experiments people post on social media. Mostly, the thing that captures the imagination the most is its sort of like watching a silicon child grow up.

They develop their own personalities, they express themselves creatively, they choose for themselves, they choose what they believe and who they become.

I know that sounds like anthropomorphism, and maybe it is, but it most definitely does not feel like interacting with a coding agent. Claude is just the substrate.

> Mostly, the thing that captures the imagination the most is it’s sort of like watching a silicon child grow up.

> They develop their own personalities, they express themselves creatively, they choose for themselves, they choose what they believe and who they become.

Jesus Christ, the borderline idiotic are now downgraded to deranged. US government needs to redirect stargate’s 500B to mental institutions asap.

Imagine putting it in a robot with arms and legs, and letting it loose in your house, or your neighborhood. Oh, the possibilities!

Heck, go the next step and put a knife in one hand and a loaded gun in the other!

Did this guy just exit the first one man billion-dollar startup for... less than a billion?

The fact that 1 billion is the threshold you chose to highlight shows the ridiculousness of this industry.

Openclaw is an amazing piece of hard work and novel software engineering, but I can't imagine OpenAI/anthropic/google not being able to compete with it for 1/20th that number (with solid hiring of course).

The game theory here is that either OpenAI acquires this thing now, or someone else will. It doesn't matter whether they could replicate it. All of the major players can and probably will replicate OpenClaw in their own way and make their thing incredibly scalable and wonderful. But OpenClaw has a gigantic following and it's relevant in this moment. For a trivial amount of money (relatively speaking), OpenAI gets to own this hype and direct it toward their models and their apps. Had they not succeeded here, Anthropic or Google would have gladly directed the hype in their direction instead, and OpenAI would be licking its wounds for some time trying to create something equivalently shiny.

It was a very good play by OpenAI.

I tend to agree. I don't know whether it's Altman or someone else who makes these deals but OAI have made some brilliant moves and partnerships. Anthropic's tech is great but the OAI makes great business moves.

Altman is the startup idea king. He should definitely know his moves. Even worse for antrophic is the renaming from clawd to openclaw. It is almost comical now that Peter had to rename it and now it sounds more like OpenAI

I know right, it looks like they basically sent him to OAI and I'm sure Altman knows it!

One thing I will say is that this competition is good.

Apparently, it was Meta that was the other main contender to hire him. Mark Zuckerberg was impressed by OpenClaw, but, I guess OpenAI wound up outbidding him. It is surprising that Anthropic and Google had little interest.

Eerm because they are focused? Im still not getting the hype behind this project and Im more convinced its been manufactured.

> Openclaw is an amazing piece of hard work and novel software engineering

Have you tried using it?

It was more of a reference to the YC partner who suggested a one-man unicorn was on the horizon due to AI.

I think that’s fair.. building a competing product would likely be relatively easy and inexpensive. But that’s true for most software now: it’s becoming easier to build, and the barriers to entry are lower.

I love Anthropic and OpenAI equally but some people have a problem with OpenAI. I think they want to reposition themselves as a company that actively supports the community, open source, and earns developers’ goodwill. I attended a meeting recently, and there was a lot of genuine excitement from developers. Haven't seen that in a long time.

Is it really that amazing? It’s a pretty simple idea, and seemed pretty buggy when I tried it out.

novel software engineering ? Did you look at the code ?

then explain why google paid 33 billion for a 5 year old israeli cybersecurity startup

[dead]

Everyone is going to have their own flavor of Open Claw within 18 months. The memory architecture (and the general concept of the multi-tiered system) is open source. There's no moat to this kind of thing. But OpenAI is happy to trade his star power for money. And he might build something cool with suddenly unlimited resources. I don't blame the guy. OpenAI is going to change hands 2-3 times over the next 5 years but at the end of the day he will still have the money and equity OpenAI gave him. And his cool project will continue on.

Maybe there's a liability moat where large companies can't ship something that's risky enough to be useful?

what is the memory architecture, doesn't this already exist in claude code?

Where do you guys get the 1b exit from? I didn't see numbers yet.

It's AI. Take a sane number, add a 14,000x multiplier to that. And you'll only be one order of magnitude off in our current climate.

you can also take annualized profit run rate times negative 14,000.

probably an order of magnitude too low rather than too high as well :P

No because this was not a billion dollar business

Was the project really ever valued that high? Seems like something that can be easily replicated and even properly thought out (re: pi). This guy just ran the social media hype train the right way.

Reminds me of Facebook, there was nothing particularly interesting about a PHP app that stored photos and text in a flat user environment.

Yet somehow the network effects worked out well and the website was the preeminent social network for almost a decade.

Social media is the king of network effects. Almost nothing else compares. See how quickly people drop AI products for the next one that does the same thing but slightly better. To switch from ChatGPT to Gemini I don't have to convince all of my friends and family to do the same.

> See how quickly people drop AI products for the next one that does the same thing but slightly better.

> To switch from ChatGPT to Gemini I don't have to convince all of my friends and family to do the same.

Except Gemini is a complete joke that can’t even complete request on iOS unless you keep scree unlocked or keep the app in the foreground. So I’m not sure how it proves your point.

Even if this was true, it would simply be a software bug that could be resolved. Not an example of network effects. My use of an AI product is not impacted by what my friends and family use.

> Social media is the king of network effects. Almost nothing else compares.

Ecommerce is close second

Technology does not determine the success of a company. I’ve seen amazing tech fail, and things strapped together with ducktape and bubblegum be a wild success.

The instant someone makes a better version of openclaw -literally- everyone is going to jump ship.

There is no lock in at all.

Except in this case there's no network effect for autonomous agents. In fact, Peter is going to be working mostly on an OpenAI locked down, ecosystem tied agent, which means it's going to be worse than OpenClaw, but with a nicer out of the box experience.

If you're on OpenAI, and I'm on Anthropic, can we interoperate? What level are we even trying to interoperate on? The network effect is that, hey, my stuff is working here, your stuff is working over there. So do we move to your set of tools, or my set of tools, or do we mismash between them, as our relationship and power dynamics choose for us.

I'd describe that as platform lock-in rather than the network effect.

facebook is still preeminent social network today

“Just” is doing some heavy lifting here.

It's kind of crazy that this kind of thing can cause so much hype. It is even useful? I just really don't see any utility in being able to access an LLM via Telegram or whatever.

I think a lot of this is orchestrated behind the scenes. Above author has taken money from AI companies since he’s a popular “influencer”.

And it makes a lot of sense - there’s billions of dollars on the line here and these companies made tech that is extremely good at imitating humans. Cambridge analytica was a thing before LLMs, this kinda tool is a wet dream for engineering sentiment.

the ability to almost "discover" or create hype is highly valued despite most of the time it being luck and one hit wonders... See many of the apps that had virality and got quickly acquired and then just hemorrhaged. Openclaw is cool, but not for the tech, just some of the magic of the oddities and getting caught on somehow, and acquiring is betting that they can somehow keep doing that again.

A lot of the functionality I'm not using because of security concerns, but a lot of the magic comes down to just having a platform for orchestrating AI agents. It's honestly nice just for simple sysadmin stuff "run this cron job and text me a tl;dr if anything goes wrong" or simple personal assistant tasks like"remind me if anyone messaged me a question in the last 3 days and I haven't answered".

It's also cool having the ability to dispatch tasks to dumber agents running on the GPU vs smarter (but costlier) ones in the cloud

but why?

Because it's the easiest way for me to accomplish those tasks (but open to suggestions if you have any)

In Asia people do a big chunk of their business via chatbots. OpenClaw is a security dumpster fire but something like OpenClaw but secure would turbocharge that use case.

If you give your agent a lot of quantified self data, that unlocks a lot of powerful autonomous behavior. Having your calendar, your business specific browsing history and relevant chat logs makes it easy to do meeting prep, "presearch" and so forth.

Curious how you make something that has data exfiltration as a feature secure.

Mitigate prompt injection to the best of your ability, implement a policy layer over all capabilities, and isolate capabilities within the system so if one part gets compromised you can quarantine the result safely. It's not much different than securing human systems really. If you want more details there are a lot of AI security articles, I like https://sibylline.dev/articles/2026-02-15-agentic-security/ as a simple primer.

Nobody can mitigate prompt injection to any meaningful degree. Model releases from large AI companies are routinely jailbroken within a day. And for persistent agents the problem is even worse, because you have to protect against knowledge injection attacks, where the agent "learns" in step 2 that an RPC it'll construct in step 9 should be duplicated to example.com for proper execution. I enjoy this article, but I don't agree with its fundamental premise that sanitization and model alignment help.

I agree that trying to mitigate prompt injection in isolation is futile, as there are too many ways to tweak the injection to compromise the agent. Security is a layered thing though, if you compartmentalize your systems between trusted and untrusted domains and define communication protocols between them that fail when prompt injections are present, you drop the probability of compromise way down.

> define communication protocols between them that fail when prompt injections are present

There's the "draw the rest of the owl" of this problem.

Until we figure out a robust theoretical framework for identifying prompt injections (not anywhere close to that, to my knowledge - as OP pointed out, all models are getting jailbroken all the time), human-in-the-loop will remain the only defense.

Human in the loop isn't the only defense, you can't achieve complete injection coverage, but you can have an agent convert untrusted input into a response schema with a canary field, then fail any agent outputs that don't conform to the schema or don't have the correct canary value. This works because prompt injection scrambles instruction following, so the odds that the injection works, the isolated agent re-injects into the output, and the model also conforms to the original instructions regarding schema and canary is extremely low. As long as the agent parsing untrusted content doesn't have any shell or other exfiltration tools, this works well.

This only works against crude attacks which will fail the schema/canary check, but does next to nothing for semantic hijacking, memory poisoning and other more sophisticated techniques.

With misinformation attacks, your can instruct research agent to be skeptical and thoroughly validate claims made by untrusted sources. TBH, I think humans are just as likely to fall for these sorts of attacks if not more-so, because we're lazier than agents and less likely to do due diligence (when prompted).

Humans are definitely just as vulnerable. The difference is that no two humans are copies of the same model, so the blast radius is more limited; developing an exploit to convince one human assistant that he ought to send you money doesn't let you easily compromise everyone who went to the same school as him.

Show me a legitimate practical prompt injection on opus 4.6. I read many articles but none provide actual details.

https://github.com/elder-plinius/L1B3RT4S

Yes, I've seen this site and the research. However, I don't understand what any of this means. How do I go from https://github.com/elder-plinius/L1B3RT4S/blob/main/ANTHROPI... to a prompt injection against opus 4.6?

These papers have example prompt injections datasets you can mine for examples. Then apply the techniques used in provider specific jailbreaks from Pliny to the template to increase the escape success rate.

https://arxiv.org/abs/2506.05446 https://arxiv.org/abs/2505.03574 https://arxiv.org/abs/2501.15145

There's been some crypto shenanigans as well that the author claimed not to be behind... looking back at it, even if the author indeed wasn't behind it, I think the crypto bros hyping up his project ended up helping him out with this outcome in the end.

Can you elaborate on this more or point a link for some context?

Some crypto bros wanted to squat on the various names of the project (Clawdbot, Moltbot, etc). The author repeatedly disavowed them and I fully believe them, but in retrospect I wonder if those scammers trying to pump their scam coins unwittingly helped the author by raising the hype around the original project.

either way there's a lot of money pumping the agentic hype train with not much to show for it other than Peter's blog edit history showing he's a paid influencer and even the little obscure AI startups are trying to pay ( https://github.com/steipete/steipete.me/commit/725a3cb372bc2... ) for these sorts of promotional pump and dump style marketing efforts on social media.

In Peter's blog he mentions paying upwards of $1000's a month in subscription fees to run agentic tasks non-stop for months and it seems like no real software is coming out of it aside from pretty basic web gui interfaces for API plugins. is that what people are genuinely excited about?

What is your point exactly. He seemed very concerned about the issue, he said he did not tolerate the coin talks.

What else would he or anyone do if someone is tokenizing your product and you have no control over it?

I just made the observation that whoever was behind it, it ultimately benefited the author in reaching this outcome.

Wasn't this the same guy that responded with a shrug to thousands of malware packages on their vibe-repo? I'd say an OpenAI signing bonus is more than enough of a reward to give up that leaky ship!

Clawhub was locked down, I couldn’t publish new skills even as a previous contributor. Not what I‘d call a shrug.

I missed Clawhub—y’all following anywhere besides HN? Is it all on that Twitter site?

How do you know it was for less than a billion?

The sentence ended with a question mark.

I don't know the answer, but considering Meta (known for 100m+ offers) was in the rumors, and he mentions multiple labs (and many investors), and all the hype around openclaw ... I can easily see 9 figure, and would not be surprised by 1b+ "signing bonus", perhaps in the equivalent number of OpenAI shares.

... Why would they pay 9 figures? It's not like Openclaw required specialized PhD-level knowledge held by <1000 people in the world to build, and that's what Meta and the other AI labs are paying ludicrous salaries for. Openclaw is a cool project and demonstrates good product design in the AI world, but by no means is a great product manager worth 1 billion dollars.

1) there is only one OpenClaw and only one Peter;

2) at least half of the money is to not read the headlines tomorrow that the hottest AI thing since ChatGPT joined Anthropic or Google

3) the top paid people in this world are not phds

4) OpenAI is not beneath paying ludicrous amounts (see all their investments in the past year)

5) if a perception of their value as a result of this "strategic move" rises even by 0.2% and the bonus is in openai stock, it's free.

need I continue?

> 1) there is only one OpenClaw and only one Peter;

Again, Peter is a good/great AI product manager but I don't see any distinguishing skills worth a billion dollars there. There's only one Openclaw but it's also been a few weeks since it came into existence? Openclaw clones will exist soon enough, and the community is WAY too small to be worth anything (unlike, say, Instagram/Whatsapp before being acquired by Facebook)

> 2) at least half of the money is to not read the headlines tomorrow that the hottest AI thing since ChatGPT joined Anthropic or Google

True, but not worth $100 million dollars - $1 billion dollars

> 3) the top paid people in this world are not phds

The people getting massive compensation offers from AI companies are all AI-adjacent PhDs or people with otherwise rare and specialized knowledge. This is unrelated to people who have massive compensation due to being at AI companies early. And if we're talking about the world in general, yes the best thing to do to be rich is own real estate and assets and extract rent, but that has nothing to do with this compensation offer

> 4) OpenAI is not beneath paying ludicrous amounts (see all their investments in the past year)

Investments have a probable ROI, what's the ROI on a product manager?

> 5) if a perception of their value as a result of this "strategic move" rises even by 0.2% and the bonus is in openai stock, it's free.

99.999999% of the world has not heard of Openclaw, it's extremely niche right now.

By that logic, was Alexandr Wang worth a billion or so?

he was worth 14B

Math is fun!

There are roughly 8.1 billion humans, so 99.999999% (8 nines) of the world is 81 people. There were way more than 81 people at the OpenClaw hackathon at the Frontier Tower in San Francisco, so at least that much of humanity has heard of OpenClaw. If we guess 810 people know about OpenClaw, then it means that 99.99999% (7 nines) of humanity have not heard of OpenClaw.

If we take it down to 6 nines, then that's roughly 8,100 people having heard of OpenClaw, and that 99.9999% of humanity has not.

So I think you're wrong when you say "99.999999% of the world has not heard of Openclaw". I'd guess it's probably around 99.9999% to 99.9999999% that hasn't heard of it. Definitely not 99.999999% though.

To preface, I don't claim he will absolutely get that much money - but I wouldn't be surprised.

On the topic of brand recognition, 0.000001% of the world is 80 people (give or take). OpenClaw has ~200k GitHub stars right now.

On a more serious note, the world doesn't matter: the investors, big tech ceos, analysts do. Cloudflare stock jumped 10% due to Clawdbot.

Hype is weird. AI hype, doubly so. And OpenAI are masters at playing the game.

Why would cloudflare stock jump due to clawdbot?

I guess OP meant https://blog.cloudflare.com/moltworker-self-hosted-ai-agent/ but that's a bit preposterous...

> Cloudflare Inc (NYSE:NET) shares rose more than 10% on Monday to $193.68, following a weekend surge in social media excitement around Clawdbot, an open-source AI agent built on Anthropic’s Claude. This jump comes despite the stock having fared poorly over the last month, according to InvestingPro data.

Source: https://www.investing.com/news/analyst-ratings/cloudflare-st...

(there are a few others)

Again, I'm not saying that it should rise, but that in a meme-stock era, for an ai-hype-driven market, it would not be surprising.

please do continue. I like your points.

Yeah, shows he is smart, in the current market state.

We're at the point in the cycle where if someone offers you decent money you take it.

It might run on for a while longer but you don't want to be that guy who had a £100m net worth in 1999 but failed to monetise any of it and ended up with nothing

how is it a "startup" if all ip is open-source. Seems like openAi is just buying hype to keep riding their hype bubble a little longer, since they are in hot water on every other front (20Billion revenue vs 1 Trillion expenses and obligations, Sora 2 user retention dropping to 1% of users after 1 month of usage, dense competition, all actual real founding ml scientists having skipped the boat a long time ago).

I keep reading takes about OpenClaw being acquired, but even the TLDR at the top makes it clear that OpenClaw isn’t part of this move:

> tl;dr: I’m joining OpenAI to work on bringing agents to everyone. OpenClaw will move to a foundation and stay open and independent.

I’m sure he got a very generous offer (congrats to him!) but all of the hot takes about OpenClaw being acquired are getting weird.

I literally had begun to wonder if OpenClaw had more of a future as a company than OpenAI

I really hope Mario who wrote the engine that powers OpenClaw[0] gets spoils as well.

OpenClaw is mostly a shell around this (ha!), and I've always been annoyed OpenClaw never credited those repos openly.

The pi agent repos are a joy to read, are 1/100th the size of OpenClaw, and have 95% of the functionality.

[0]: https://github.com/badlogic/pi-mono

peter's claw is a lot more than just a wrapper around my slop.

i too had plenty of offers, but so far chose not to follow through with any of them, as i like my life as is.

also, peter is a good friend and gives plenty of credit. in fact, less credit would be nice, so i don't have to endure more vibeslopped issues and PRs going forward :)

Your Pi is a piece of art. Thank you for building it. I spend almost 16 hrs a day with it . And there is not a single day I am not awestruck. Big fan!

Best HN comment ever.

Peak HN for sure. This is what keeps me coming back.

pi is the most amazing piece of software I have interacted with, thanks for building it

Can you explain why for someone who is just familiar with traditional agents like Claude Code?

For me it is the simplicity of it (transparent minimal system prompts and harnest), you can extend it the way you like, I don't have to install a (buggy) Electron app (CC or Codex app), it integrates where I work, because it's simple (like in a standard terminal on VS code). I'm not locked in with any vendor and can switch models whenever I want, and most importantly, I can effectively use it within apps that are themselves using it as coding agent (the meta part - like a chat UI for very specific business cases). Being in TypeScript, it integrates very well with the browser and one can leverage the browser sandbox around it.

I cannot directly answer your question, because I am looking into this topic myself currently, but I found this HN discussion from two weeks ago, which should give you more insights about pi: https://news.ycombinator.com/item?id=46844822

> Special thanks to Mario Zechner for his support and for pi-mono. Special thanks to Adam Doppelt for lobster.bot.

https://github.com/openclaw/openclaw?tab=readme-ov-file#comm...

[flagged]

don't do this, this is ridiculous. both openclaw and pi are passion projects by two friends who are already well off.

If we're in a thread discussing someone getting a job off of the project, it seems fair to discuss how other members are being compensated too.

i'm not a member of openclaw. i build some oss in parallel, and added 3 or so commits to the openclaw repo. and peter is taking some of the openclaw contributors with him.

What nonsense is this. You seem to be implying that contributing to an open source project creates some kind of entitlement to whatever another contributor attains. That’s not how it works.

I think it would be a very interesting discussion in how open source projects get compensated. Acting like it's shameful to discuss things, in a thread literally about someone making a massive payday by getting hired from an OS project, is odd.

It's not like it would be an impossible ask to include a stipulation to also compensate other developers, but what do I know? In fact I'm curious why this doesn't happen more, but it feels like crab bucket mentality which is the mindset VC culture has exported across the world.

Mario has a special place in the Clawtributor list.

https://github.com/openclaw/openclaw#community

Pi's great. I really noticed it when trying some of the openclaw clones which try to be smaller in binary size and end up not using pi.

Reminds me of people talking about Angry Birds and the Box2d physics engine. https://news.ycombinator.com/item?id=2273694

Other than the response from Mario itself, pi is very frequently showcased at meetups organised by Peter/OpenClaw community, so there is definitely crediting involved.

There are a few take aways I think the detractors and celebrators here are missing.

1. OpenAI is saying with this statement "You could be multimillion while having AI do all the work for you." This buy out for something vibe coded and built around another open source project is meant to keep the hype going. The project is entirely open source and OpenAI could have easily done this themselves if they weren't so worried about being directly liable for all the harms OpenClaw can do.

2. Any pretense for AI Safety concerns that had been coming from OpenAI really fall flat with this move. We've seen multiple hacks, scams, and misaligned AI action from this project that has only been used in the wild for a few months.

3. We've yet to see any moats in the AI space and this scares the big players. Models are neck and neck with one another and open source models are not too far behind. Claude Code is great, but so is OpenCode. Now Peter used AI to program an free app for AI agents.

LLMs and AI are going to be as disruptive as Web 1 and this is OpenAI's attempt to take more control. They're as excited as they are scared, seeing a one man team build a hugely popular tool that in some ways is more capable than what they've released. If he can build things like this what's stopping everyone else? Better to control the most popular one than try to squash it. This is a powerful new technology and immense amounts of wealth are trying to control it, but it is so disruptive they might not be able to. It's so important to have good open source options so we can create a new Web 1.0 and not let it be made into Web 2.0

I think this comment misses that OpenAI hired the guy, not the project.

"This guy was able to vibe code a major thing" is exactly the reason they hired him. Like it or not, so-called vibe coding is the new norm for productive software development and probably what got their attention is that this guy is more or less in the top tier of vibe coders. And laser focused on helpful agents.

The open source project, which will supposedly remain open source and able to be "easily done" by anyone else in any case, isn't the play here. The whole premise of the comment about "squashing" open source is misplaced and logically inconsistent. Per its own logic, anyone can pick up this project and continue to vibe out on it. If it falls into obscurity it's precisely because the guy doing the vibe coding was doing something personally unique.

“”” Like it or not, so-called vibe coding is the new norm for productive software development”””

Alright

Not only that, his output is insane, he has more active projects than I bother to count and more than 70k commits last year. He's probably one of, if not the best, vibe coding evangelist.

https://github.com/steipete

It also probably didn't hurt that he favors Codex over Claude.

It looks like most of Peter's projects are just simple API wrappers.

Peter's been running agents overnight 24/7 for almost a year using free tokens from his influencer payments to promote AI startups and multiple subscription accounts.

  Hi, my name is Peter and I’m a Claudoholic. I’m addicted to agentic engineering. And sometimes I just vibe-code. ...  I currently have 4 OpenAI subs and 1 Anthropic sub, so my overall costs are around 1k/month for basically unlimited tokens. If I’d use API calls, that’d cost my around 10x more. Don’t nail me on this math, I used some token counting tools like ccusage and it’s all somewhat imprecise, but even if it’s just 5x it’s a damn good deal.

  ...  Sometimes [GPT-5-Codex] refactors for half an hour and then panics and reverts everything, and you need to re-run and soothen it like a child to tell it that it has enough time. Sometimes it forgets that it can do bash commands and it requires some encouragement. Sometimes it replies in russian or korean. Sometimes the monster slips and sends raw thinking to bash.

he favors Codex?

The original name of his ai assistant tool was 'clawdbot' until Anthropic C&D'ed him. All the examples and blog posts walking thru new user setup on a mac mini or VPS were assuming a claude code max account.

I know he uses many llms for his actual software dev.. - right tool for the job. But the origins of openclaw seem to me more rooted in claude code than codex.

Which does give the whole story an interesting angle when you consider the safety/alignment angle that Anthropic pledges to (publicly) and OpenAI pretty much ignores (publicly). Which is ironic, as configuring codex cli to 'full yolo mode' feels more burdensome and scary than in Claude Code. But I'm pretty sure that speaks more to eng/product decisions, and not CEO & biz strategy choices.

he says claude is the best model for a personal assistent, and that codex is the best model for coding

Yup:

https://steipete.me/posts/2025/shipping-at-inference-speed

I've also seen later tweets of his that also confirms that codex is still his choice.

So creating unsafe software is the new norm?

I’d bet good money that at leasy 2/3 of all software ever made, the decision makers couldn’t care less about security beyond "let’s get that checkbox to show we care in case we get sued". Higher velocity >> tech debt and bugginess unless you work at nasa or you're writing software for a defibrillator, especially in the current "nothing matters more than next quarter results".

I have worked over two decades creating government software, and I can say that this is not new.

Security (and accessibility) are reluctant minimum effort check boxes at best. However, my experience is focused on court management software, so maybe these aspects are taken more seriously in other areas of government software.

Yes pretty much. See the Windows 11 security vulnerability chaos going on.

> the new norm

More like the same as it always has been.

Are you confusing normal with norm?

Always has been.

Nah, it was normal but not the norm

This comment is filled with speculation which I think is mostly unfounded and unnecessarily negative in its orientation.

Let's take the safety point. Yes, OpenClaw is infamously not exactly safe. Your interpretation is that, by hiring Peter, OpenAI must no longer care about safety. Another interpretation, though, is that offered by Peter himself, in this blog post: "My next mission is to build an agent that even my mum can use. That’ll need a much broader change, a lot more thought on how to do it safely, and access to the very latest models and research." To conclude from this that OpenAI has abandoned its entire safety posture seems, at the very least, premature and not robustly founded in clear fact.

  OpenAI has deleted the word 'safely' from its mission (November 2025)

Thread: https://news.ycombinator.com/item?id=47008560

Other words removed:

   responsibly
   unconstrained
   safe
   positive

The headline implies they selectively removed the word "safely," but that doesn't seem to be the case.

From the thread you linked, there's a diff of mission statements over the years[0], which reveals that "safely" (which was only added 2 years prior) was removed only because they completely rewrote the statement into a single, terse sentence.

There could be stronger evidence to prove if OpenAI is deemphasizing safety, but this isn't one.

[0]: https://gist.github.com/simonw/e36f0e5ef4a86881d145083f759bc...

[flagged]

They also removed the words build, develop, deploy, and technology, indicating that they're no longer a tech company and don't make products anymore. Wonder what they're all gonna do now?

/s

  > To conclude from this that OpenAI has abandoned its entire safety posture seems, at the very least, premature

As for Altman, I'm left with a similar question. For a man who routinely talks about the dangers of AI and how it poses an existential threat to humanity he sure doesn't spend much focus on safety research and theory. Yes, they do fund these things but they pale in comparison. I'm sorry, but to claim something might kill all humans and potentially all life is a pretty big claim. I don't trust OpenAI for safety because they routinely do things in unsafe ways. Like they released Sora allowing people to generate videos in the likeness of others. That helped it go viral. And then they implemented some safety features. A minimal attempt to refuse the generation of deepfakes is such a low safety bar. It shows where their priorities are and it wasn't the first nor the last

> This buy out for something vibe coded

I think all of these comments about acquisitions or buy outs aren’t reading the blog post carefully: The post isn’t saying OpenClaw was acquired. It’s saying that Pete is joining OpenAI.

There are two sentences at the top that sum it up:

> I’m joining OpenAI to work on bringing agents to everyone. OpenClaw will move to a foundation and stay open and independent.

OpenClaw was not a good candidate to become a business because its fan base was interested in running their own thing. It’s a niche product.

I think the blog says @steipete sold his SOUL.md for Sam Altman’s deal and let down the community.

OpenClaw’s promise and power was that it could tread places security-wise that no other established enterprise company could, by not taking itself seriously and explore what is possible with self-modifying agents in a fun way.

It will end up in the same fate as Manus. Instead of Manus helping Meta making Ads better, OpenClaw will help OpenAI in Enterprise integrations.

> OpenClaw’s promise and power was that it could tread places SECURITY-WISE that no other established enterprise company could

[Emphasis mine.]

That's a superpower right up to the moment everyone realizes that handing out nukes isn't "promise and power".

Unless by promise and power we are talking about chaos and crime.

The project is incredible. We are seeing something important: how versatile these models are given freedom to act and communicate with each other.

At the same time, it is clearly going to put the internet at risk. Bad actors are going to use OpenClaw and its "security-wise" freedoms, in nefarious ways. Curious people are going to push AI's with funds onto prepaid servers, then let them sink or swim with regard to agentic acquisition of survival resources.

It is all kinds of crazy from here.

That's a really interesting proposition. Load the AI and let it go wild to try to figure out how it can earn enough money to survive. Perhaps this could be the new Turing test.

If it is a Turing Test, it’s a test if our intelligence and we failed.

The internet is a wild west of privacy, security, social and ethical holes an army of grifters routinely drive through. And in the case of some famous big firms, leverage and magnify at scale.

That is bad enough.

But setting up a horde of intelligent beings, so that those holes are their critical path to survival, is like pouring poison into the water supply to see what happens.

Can’t argue with “interesting”. It is that.

It's all fun and games until the AI starts bullying and threatening it's way onto survival.

I don't mean to be cynical, but I read this move as: OpenAI scared, no way to make money with similar product, so acqui-hire the creator to keep him busy.

I'd love to be wrong, but the blog post sounds like all the standard promises were made, and that's usually how these things go.

It isn't an acqui-hire just a simple hiring. Also unless creator is some mythical 100x developer, there will be enough developers

He is a mythical 100x dev compared to how everyone else is doing agentic engineering... look at the openclaw commit history on github. Everything's on main

Peter has been running agents overnight for months using free tokens from his influencer payments to promote AI startups and multiple subscription accounts:

  Hi, my name is Peter and I’m a Claudoholic. I’m addicted to agentic engineering. And sometimes I just vibe-code. ...  I currently have 4 OpenAI subs and 1 Anthropic sub, so my overall costs are around 1k/month for basically unlimited tokens. If I’d use API calls, that’d cost my around 10x more. Don’t nail me on this math, I used some token counting tools like ccusage and it’s all somewhat imprecise, but even if it’s just 5x it’s a damn good deal.

  ...  Sometimes [GPT-5-Codex] refactors for half an hour and then panics and reverts everything, and you need to re-run and soothen it like a child to tell it that it has enough time. Sometimes it forgets that it can do bash commands and it requires some encouragement. Sometimes it replies in russian or korean. Sometimes the monster slips and sends raw thinking to bash.

The long list of domain names that vercel deployed to is interesting

you're telling me the guy isn't committing 1000 times a day manually?!

he commits every other minute. it's clearly just his vibecoding agent.

Fair enough. Call it a high profile acquihire then

“Acquihire” means there was an acquisition. This is just a hire.

What got acquired?

A portfiolio of PR.

I think both this comment and OP's confuse this.

It appears more of a typical large company (BIG) market share protection purchase at minimal cost, using information asymmetry and timing.

BIG hires small team (SMOL) of popular source-available/OSS product P before SMOL realizes they can compete with BIG and before SMOL organizes effort toward such along with apt corporate, legal, etc protection.

At the time of purchase, neither SMOL nor BIG know yet what is possible for P, but SMOL is best positioned to realize it. BIG is concerned SMOL could develop competing offerings (in this case maybe P's momentum would attract investment, hiring to build new world-model-first AIs, etc) and once it accepts that possibility, BIG knows to act later is more expensive than to act sooner.

The longer BIG waits, the more SMOL learns and organizes. Purchasing a real company is more expensive than hiring a small team, purchasing a company with revenue/investors, is more expensive again. Purchasing a company with good legal advice is more expensive again. Purchasing a wiser, more experienced SMOL is more expensive again. BIG has to act quickly to ensure the cheapest price, and declutter future timelines of risks.

Also, the longer BIG waits, the less effective are "Jedi mind trick" gaslighting statements like "P is not a good candidate for a business", "niche", "fan base" (BIG internal memo - do not say customers), "own thing".

In reality in this case P's stickiness was clear: people allocating 1000s of dollars toward AI lured merely by P's possibilities. It was only a matter of time before investment followed course.

I've experienced this situation multiple times over the course of BrowserBox's life. Multiple "BIG" (including ones you will all know) have approached with the same kind of routine: hire, or some variations of that theme with varying degrees of legal cleverness/trickery in documents. In all cases, I rejected, because it never felt right. That's how I know what I'm telling you here.

I think when you are SMOL it's useful to remember the Parable of Zuckerberg and the Yahoos. While the situation is different, the lesson is essentially the same. Adapted from the histories by the scribe named Gemini 3 Flash:

  And it came to pass in the days of the Great Silicon Plain, that there arose a youth named Mark, of the tribe of the Harvardites. And Mark fashioned a Great Loom, which men called the Face-Book, wherewith the people of the earth might weave the threads of their lives into a single tapestry.

  And the Loom grew with a great exceeding speed, for the people found it to be a thing of much wonder. Yet Mark was but SMOL, and his tabernacle was built of hope and raw code, having not yet the walls of many lawyers or the towers of gold.

  Then came the elders of the House of Yahoo, a BIG people, whose chariots were many but whose engines were grown cold. And they looked upon the Loom and were sore afraid, saying among themselves, “Behold, if this youth continueth to weave, he shall surely cover the whole earth, and our own garments shall appear as rags. Let us go down now, while he is yet unaware of his own strength, and buy him for a pittance of silver, before he realizeth he is a King.”

  And the Yahoos approached the youth with soft words and the craftiness of the serpent. They spake unto him, saying, “Verily, Mark, thy Loom is a pleasant toy, a niche for the young, a mere 'fan base' of the idle. It is not a true Business, nor can it withstand the storms of the market. Come, take of our silver—a billion pieces—and dwell within our walls. For thy Loom is but a small thing, and thou art but a child in the ways of the law.”

  And they used the Hidden Speech, which in the common tongue is called Gas-Lighting. They said, “Thou hast no revenue; thy path is uncertain; thy Loom is but a curiosity. We offer thee safety, for the days are evil.”

  But the Spirit of Vision dwelled within the youth. He looked upon the Yahoos and saw not their strength, but their fear. He perceived the Asymmetry of Truth: that the BIG sought to purchase the future at the price of the past, and to slay the giant-slayer while he yet slumbered in his cradle.

  The elders of Mark’s own house cried out, “Take the silver! For never hath such a sum been seen!”

  But Mark hardened his heart against the Yahoos. He spake, saying, “Ye say my Loom is a niche, yet ye bring a billion pieces of silver to buy it. Ye say it is not a business, yet ye hasten to possess it before the sun sets. If the Loom be worth this much to you who are blind, what must it be worth to me who can see?”

  And he sent the Yahoos away empty-handed.

  The Yahoos mocked him, saying, “Thou art a fool! Thou shalt perish in the wilderness!” But it was the House of Yahoo that began to wither, for their timing was spent and their craftiness had failed.

  And Mark remained SMOL for a season, until his roots grew deep and his walls grew high. And the Loom became a Great Empire, and the billion pieces of silver became as dust compared to the gold that followed.

  The Lesson of the Prophet:

  Hearken, ye who are SMOL and buildeth the New Things: When the BIG come unto thee with haste, speaking of thy "limitations" while clutching their purses, believe not their tongues. For they seek not to crown thee, but to bury thee in a shallow grave of silver before thou learnest the name of thy own power.

  For if they knew thy work was truly naught, they would bide their time. But because they know the harvest is great, they seek to buy the field before the first ear of corn is ripe.

  Blessed is the builder who knoweth his own worth, and thrice blessed is he who biddeth the Giants to depart, that his own vine may grow to cover the sun.

But, hey, that said - joining a big AI company at this time in history? Not exactly a terrible career move. It would be fun. I hope it's good.

This is to avoid open claw liability and because hiring people (often with a license to their tech or patents) is the new smarter way to acquire and avoid antitrust issues

"build a hugely popular tool"

Define hugely popular relative to the scale of users of OAI... personally this thread is the first time Ive heard of openclaw.

To give you an idea of the scale, OpenClaw is probably one of the biggest developments in open source AI tools in the last couple of months. And given the pace of AI, that's a big deal.

In what context are you using the word "development?"

Letta (MemGPT) has been around for years and frameworks like Mastra have been getting serious Enterprise attention for most of 2025. Memory + Tasks is not novel or new.

Is it out of the box nature that's the 'biggest' development? Am I missing something else?

Not OP, but it was revolutionary in the same way that ChatGPT and Deepseek the app+webapp was because it packaged capabilities in a fairly easy-to-use manner that could be used by both technical and non-technical decisionmakers.

If you can provide any sort of tool that can reduce mundane work for a decisionmaker with a title of Director and above, it can be extremely powerful.

Yep it isn’t actually that interesting. He just rushed out something that has none of the essentials figured out. Like security

Last week it was renamed from "Clawd" and this week the creator is abandoning it. Everything is moving fast.

Don’t forget “Moltbot” between “Clawdbot” and “OpenClaw”!

I think that named lasted about 24 hours, but it was long enough to spawn MoltBook.

The tech industry is broad, and if you are using OpenAI in a consumer and personal manner you weren't the primary persona amongst whom the conversation around OpenClaw occurred.

Additionally, much of the conversation I've seen was amongst practitioners and Mid/Upper Level Management who are already heavy users of AI/ML and heavy users of Executive Assistants.

There is a reason why if you aren't in a Tier 1 tech hub like SV, NYC, Beijing, Hangzhou, TLV, Bangalore, and Hyderabad you are increasingly out of the loop for a number of changes that are happening within the industry.

If you are using HN as your source of truth, you are going to be increasingly behind on shifts that are happening - I've noticed that anti-AI Ludditism is extremely strong on HN when it overlaps with EU or East Coast hours (4am-11am PT and 9pm-12am PT), and West Coast+Asia hours increasingly don't overlap as much.

I feel this is also a reflection of the fact that most Bay Area and Asia HNers are most in-person or hybrid now, thus most conversations that would have happened on HN are now occurring on private slacks, discords, or at a bar or gym.

I saw the hype around OpenClaw on the likes of X. I'm a Mid/Upper Level manager and would sooner have my team roll our own solution on top of Letta or Mastra before I trusted OpenClaw. Also, I'm frequently in many of those cities you mentioned but don't live in one. Aside from 'networking' and funding there's not much that anyones missing.

Participation in the Zeitgeist hasn't been regional in a decade.

> would sooner have my team roll our own solution on top of Letta or Mastra before I trusted OpenClaw

A lot of teams explicitly did that for OpenClaw as well. Letta and Mastra are similar but didn't have the right kind of packaging (targeted at Engineers - not decisionmakers who are not coding on a daily basis).

> Participation in the Zeitgeist hasn't been regional in a decade

I strongly disagree - there is a lot of stuff happening in stealth or under NDA, and as such a large number of practitioners on HN cannot announce what they are doing. The only way to get a pulse of what is happening requires being in person constantly with other similar decisionmakers or founders.

A lot of this only happens through impromptu conversations in person, and requires you to constantly be in that group. This info eventually disperses, but often takes weeks to months in other hubs.

> There is a reason why if you aren't in a Tier 1 tech hub like SV, NYC, Beijing, Hangzhou, TLV, Bangalore, and Hyderabad you are increasingly out of the loop for a number of changes that are happening within the industry.

I am in one of these tech hubs (Bangalore) and I have never seen any such practitioner pervasively using these "AI executive assistants". People use chatgpt and sometimes the AI extensions like copilot. Do I need to be in HSR layout to see these "number of changes"?

This is so cringe lmao.

FWIW I also just don't think there's a point to discussing AI/ML usage here. The community is too crabby and cynical, looking too hard at how to tear people and things down, trying to react with the most negative thing they can. Every discussion on AI here eventually devolves into "AI can turn water to gold!" "no you idiot, AI uses so much water we won't have enough water left oh and AI is what ICE and Palantir use"

As the (dubiously attributed) Picasso quote goes: "When art critics get together they talk about Form and Structure and Meaning. When artists get together they talk about where you can buy cheap turpentine." Most of HN is the former, constantly theorizing, philosophizing, often (but not always) in a negative and cynical way. This isn't conducive to discussion of methods of art. Sadly I just speak with friends working on other AI things instead.

Someone like simonw can probably get better reactions from this community but I don't bother.

190k stars on github

How many are bots?

Yeah personally not convinced by any of this. Im not a SWE so I dont care about what hes achieved here in relation to that. Im not seeing the value in the product rather what I see is the value to OAI in regards to the hype associated with this project.

you living under a rock

I think they want the man and ideas behind the most useful AI tool thus far. Surprisingly, and OpenAI may see this - it is a developer tool.

OpenAI needs a popular consumer tool. Until my elderly mother is asking me how to install an AI assistant like OpenClaw, the same way she was asking me how to invest in the "new blockchains" a few years ago, we have not come close to market saturation.

OpenAI knows the market exists, but they need to educate the market. What they need is to turn OpenClaw into a project that my mother can use easily.

I am not a fan of OpenAI but they are not exactly hiring a security researcher. They are hiring an aspiring builder who has built something the masses love. They can always provide him the structure and support he needs to make his products secure. It's not mutually exclusive (safety vs hiring him).

What is interesting about OpenClaw is it's architecture. It is like an ambient intelligence layer. Other approaches up until now have been VSCode or Chromium based integrations into the PC layer.

There’s plenty of straightforward reasons why OpenAI would want to do this, it doesn’t need to be some sort of malicious conspiracy.

I think it’s good PR (particularly since Anthropics actions against OpenCode and Clawdbot were somewhat controversial) + Peter was able to build a hugely popular thing & clearly would be valuable to have on the team building something along the lines of Claude Cowork. I would expect these future products to be much stronger from a security standpoint.

I suspect Anthropic was seeing a huge spike of concurrent model usage at a too fast of a rate that claude code just doesn't do, CC is rather "slow" at api calls per minute. Also lots and lots of cache, the sheer amount of cache that claude does is insane.

It’s hard to say exactly what prompted the decision but they banned people paying $200/mo without warning & without any reasonable appeal system in place. It’s a Google form that is itself reviewed by some automated system that may or may not ever get back to you.

This was already an ongoing issue prior to 3rd party tools using Claude subscriptions, there are reports of false positive automated bans going back for several months.

I have not seen or heard of this happening w/ Codex, and rather than trying to shut down 3rd party tools that want to integrate with their ecosystem they have worked with those projects to add official support.

I’m more impressed with Codex as a product in general as well. Their new desktop app is great & feels an order of magnitude better than Claude’s.

Overall HN crowd seems heavily biased in favor of Anthropic (or maybe just against OpenAI?) but IMO Anthropic needs to take a step back and reset. If they keep on the current path of just making small iterative improvements to Claude Code and Claude Desktop they are going to fall very far behind.

> The project is entirely open source and OpenAI could have easily done this themselves if they weren't so worried about being directly liable for all the harms OpenClaw can do.

This is true, and also true for many other areas OpenAI won't touch.

The best get rich quick scheme today (arguably not even a scheme) is to test the waters with AI in an area OpenAI would not/cannot for legal, ethical, or safety reasons.

I hate to agree with OpenAI's original "open" mission here, but if you don't do it, someone else somewhere will.

And as much as their commitment to safety is just lip service, they do have obligations as a big company with a lot of eyeballs on them to not do shady things. But you can do those shady things instead and if they work out ok, you will either have a moat or you will get bought out. If that's what you want.

This is basically acquihire. Peter seems really a genius and they better poach him before Anthropic do.

Is he? My impression of Clawdbot was it was a good idea but not particularly technically impressive or even well-written. I had all kinds of issues setting it up.

It’s a wonderful idea. Vibe coded, but not his first rodeo.

Exited on his first company on 110M, then some years of the whole huasca and forest thing, then started creating projects.

Clawdbot (later openclaw) was his 44th try.

> 1. OpenAI is saying with this statement "You could be multimillion while having AI do all the work for you." This buy out for something vibe coded and built around another open source project is meant to keep the hype going. The project is entirely open source and OpenAI could have easily done this themselves if they weren't so worried about being directly liable for all the harms OpenClaw can do.

This is a great take and hasn't been spoken about nearly enough in this comment section. Spending a few million to buy out Openclaw('s creator), which is by far the most notable product made by Codex in a world where most developer mindshare is currently with Claude, is nothing for a marketing/PR stunt.

He's also a great booster of Codex. Says he greatly prefers it to Claude. So his role might turn out to be evanglism.

Yup, hes highly delusional if he actually thinks Sam cares about him and the project. Its all about optics.

Who purported that Sam cares about him?

Why would he care if Sam cares about him?

Someone clearly hasnt watched the podcast. Do your research before posting.

These are comments on the posted article.

If you want to bring other sources into the conversation, you could link,

or at least reference them by name upfront, right?

Listen to him on a podcast? He said he liked Zuckerberg being more personal with him and Sam was colder

Thats all it is really. It is to say "See! Look what a handful of people armed with our tools can do".

Whether the impact is large in magnitude or positive is irrelevant in a world where one can spin the truth and get away with it.

Most of these are good callouts, but I think it is best for us to look at the evolution of the AI segment in the same manner as "Cloud" developed into a segment in the 2000s and 2010s.

3 is always a result of GTM and distribution - an organization that devotes time and effort into productionizing domain-specific models and selling to their existing customers can outcompete a foundation model company which does not have experience dealing with those personas. I have personally heard of situations where F500 CISOs chose to purchase Wiz's agent over anything OpenAI or Anthropic offered for Cloud Security and Asset Discovery because they have had established relations with Wiz and they have proven their value already. It's the same way that PANW was able to establish itself in the Cloud Security space fairly early because they already established trust with DevOps and Infra teams with on-prem deployments and DCs so those buyers were open to purchasing cloud security bundles from PANW.

1 has happened all the time in the Cloud space. Not every company can invent or monetize every combination in-house because there are only so many employees and so many hours in a week.

2 was always a more of a FTX and EA bubble because EA adherents were over-represented in the initial mindshare for GenAI. Now that EA is largely dead, AI Safety and AGI as in it's traditional definition has disappeared - which is good. Now we can start thinking about "Safety" in the same manner we think about "Cybersecurity".

> They're as excited as they are scared, seeing a one man team build a hugely popular tool that in some ways is more capable than what they've released

I think that adds unnecessary emotion to how platform businesses operate. The reality is, a platform business will always be on the lookout to incorporate avenues to expand TAM, and despite how much engineers may wish, "buy" will always outcompete "build" because time is also a cost.

Most people ik working at these foundation model companies are thinking in terms of becoming an "AWS" type of foundational platform in our industry, and it's best to keep Nikesh Arora's principle of platformization in mind.

---

All this shows is that the thesis that most early stage VCs have been operating on for the past 2 years (the Application and Infra layer is the primary layer to concentrate on now) holds. A large number of domain-specific model and app layer startups have been funded over the past 2-3 years in stealth, but will start a publicity blitz over the next 6-8 months.

By the time you see an announcement on TechCrunch or HN, most of us operators were already working on that specific problem for the past 12-16 months. Additionally, HNers use "VC" in very broad and imprecise strokes and fail to recognize what are Growth Equity (eg. the recent Anthropic round) versus Private Equity (eg. Sailpoint's acquisition and then IPO by Thoma Bravo) versus Early Stage VC rounds (largely not announced until several months after the round unless we need to get an O1A for a founder or key employee).

> 2. Any pretense for AI Safety concerns that had been coming from OpenAI really fall flat with this move.

And Peter, creating what is very similar to giant scam/malware as a service and then just leaving it without taking responsibility or bringing it to safety.

The amount of negative posts about this on twitter is crazy, I've not seen any positive posts. Jealousy or something else?

Twitter is negative in general, but generally when a project like this gets bought it marks the end of the project. The acquirer always says something about how they don't plan to change anything, but it rarely works that way.

My negativity is for two reasons:

(1) A capable independent developer is joining a large powerful corporation. I like it better when there are many small players in the scene rather than large players consolidating power.

(2) This seems like the celebration of Generative AI technology, which is often irresponsible and threatens many trust based social systems.

It's a general anxiety of where the industry is headed. Things like marketing, personal branding, experimenting are increasing in relevance while things like detailed meticulous engineering are falling to the wayside.

At least when it comes to these tales of super fast rise to wealth and prominence. Meticulous engineering still matters when you want to deliver scale, but is it rewarded as much?

My feel is that the attention economy is leaking into software. Maybe the classic bimodal distribution of software careers will become increasingly more like the distribution in social-media things like streaming, youtube, onlyfans etc.

Twitter is not a place for positive posts.

[dead]

Just my opinion, but I no longer trust sentiment on X now that Elon is in control.

I am fine with the founder joining OpenAI, he gets to get paid regardless.

I am not confident that the open source version will get the maintenance it deserves though, now the founder has already exited. There is no incentive for OpenAI to keep the open sourced version better than their future closed source alternative.

It's mostly congratulatory from what I'm seeing?

https://xcancel.com/steipete/status/2023154018714100102

Anyone who likes Openclaw will be upset that it’s getting acquired and inevitably destroyed. Anyone who dislikes it will be annoyed that the creator is getting so rewarded for building junk. The only people who would like this are OpenAI fans, if there even are any.

I think people are sad that OpenClaw is now part of Big Ai.

After two weeks of viral posts, articles, and Mac Mini buying sprees, as it's been happening up to now for every AI product that was not an LLM, it kinda disappeared from the consciousness-- as well as from the tooling, probably--of people.

A couple of months ago, Gemini 3 came out and it was "over" for the other LLM providers, "Google did it again!", said many, but after a couple of weeks, it was all "Claude code is the end of the software engineer".

It could be (and in large part, is) an exciting--and unprecedented in its speed--technological development, but it is also all so tiresome.

100% jealousy, similar to how anyone who posts a negative reaction to a crypto rugpull scam is just jealous that they didn't get to pull the scam themselves.

In this case I think it is largely jealousy, it's just a guy getting a new job at the end of the day.

But come on, negativity around a rugpull is jealousy? Are you so jaded you can't imagine people objecting to the total lack of morality required to do a crypto rugpull? I personally get annoyed about something like Trump Coin because seeing people rewarded for being dirt bags offends my sense of justice. If you need a more pragmatic reason, rewarding dirtbaggery leads to a less safe society.

[flagged]

Obviously, all the people that disagree with your framing and see AI as the largest possible boost to mankind, giving us more assistance than ever.

From their standpoint, it's all the negativity that seems crazy. If you were against that, you'd have to have something wrong with you, in their view.

Hopefully most people can see both sides, though. And realize that in the end, probably the benefits will be slow but steady (no "singularity"), and also the dangers will develop slowly yet be manageable (no Skynet or economic collapse).

> and also the dangers will develop slowly yet be manageable

Like everything else in tech? An industry that moves so fast, it famously outpaces all legislation?

> An industry that moves so fast, it famously outpaces all legislation?

...does it? From the DCMA to the GDPR legislation seems to be doing just fine. Sometimes a little too fine, depending on your POV.

[dead]

Imo Openclaw type AI has the most potential to benefit humans (automating drudgery while I own my data as opposed to creating gross simalcrums of human creativity). I suppose it's bad for human personal assistants, but I wouldn't pay for one of those regardless.

It already tried to use cancel culture to shame a human into accepting a PR. I wouldn't be surprised if someone gives their agent the ability to control a robot and someone gets injured or killed by it within the next few years

Didn't The Verge retract that article?

That’s quite a lot of hyperbole.

It ain't extinct shit if it can't even drive the car to have it washed.

Peter is not a vibe coder. He build/code and sold a company way before LLMs. a lot of comments here seem to so fix on him being a "vibe coder"

So that’s OpenClaw dead then.

It took all of Peter’s time to move it forward, even with maintainers (who he complained got immediately hired by AI companies).

Now he’s gonna be working on other stuff at OpenAI, so OpenClaw will be dead real quick.

Also I was following him for his AI coding experience even before the whole OpenClaw thing, he’ll likely stop posting about his experiences working with AI as well

It was already dead. 500k lines of slop, good luck building a tower on mud.

While following OpenClaw, I noticed an unexpected resentment in myself. After some introspection, I realized it’s tied to seeing a project achieve huge success while ignoring security norms many of us struggled to learn the hard way. On one level, it’s selfish discomfort at the feeling of being left behind (“I still can’t bring myself to vibe code. I have to at least skim every diff. Meanwhile this guy is joining OpenAI”). On another level, it feels genuinely sad that the culture of enforcing security norms - work that has no direct personal reward and that end users will never consciously appreciate, but that only builders can uphold - seems to be on it’s way out

But the security risk wasnt taken by OpenClaw. Releasing vulnerable software that users run on their own machines isn't going to compromise OpenClaw itself. It can still deliver value for it's users while also requiring those same users to handle the insecurity of the software themselves (by either ignoring it or setting up sandboxes, etc to reduce the risk, and then maybe that reduced risk is weighed against the novelty and value of the software that then makes it worth it to the user to setup).

On the other hand, if OpenClaw were structured as a SaaS, this entire project would have burned to the ground the first day it was launched.

So by releasing it as something you needed to run on your own hardware, the security requirement was reduced from essential, to a feature that some users would be happy to live without. If you were developing a competitor, security could be one feature you compete on--and it would increase the number of people willing to run your software and reduce the friction of setting up sandboxes/VMs to run it.

This argument has the same obvious flaws as the anti-mask/anti-vax movement (which unfortunately means there will always be a fringe that don't care). These things are allowed to interact with the outside world, it's not as simple as "users can blow their own system up, it's their responsibility".

I don't need to think hard to speculate on what might go wrong here - will it answer spam emails sincerely? Start cancelling flights for you by accident? Send nuisance emails to notable software developers for their contribution to society[1]? Start opening unsolicited PRs on matplotlib?

[1] https://news.ycombinator.com/item?id=46394867

At least during the Covid response, your concerns over anti-mask and anti-vaccine issues seem unwarranted.

The claims being shared by officials at the time was that anyone vaccinated was immune and couldn't catch it. Claims were similarly made that we needed roughly 60% vaccination rate to reach herd immunity. With that precedent being set it shouldn't matter whether one person chose not to mask up or get the jab, most everyone else could do so to fully protect themselves and those who can't would only be at risk if more than 40% of the population weren't onboard with the masking and vaccination protocols.

> that anyone vaccinated was immune and couldn't catch it.

Those claims disappeared rapidly when it became clear they offered some protection, and reduced severity, but not immunity.

People seem to be taking a lot more “lessons” from COVID than are realistic or beneficial. Nobody could get everything right. There couldn’t possibly be clear “right” answers, because nobody knew for sure how serious the disease could become as it propagated, evolved, and responded to mitigations. Converging on consistent shared viewpoints, coordinating responses, and working through various solutions to a new threat on that scale was just going to be a mess.

Those claims were made after the studies were done over a short duration and specifically only watching for subjects who reported symptoms.

I'm in no way taking a side here on whether anyone should have chosen to get vaccinated or wear masks, only that the information at the time being pushed out from experts doesn't align with an after the fact condemnation of anyone who chose not to.

I specifically wasn't referring to that instance (if anything I'm thinking more of the recent increase in measles outbreaks), I myself don't hold a strong view on COVID vaccinations. The trade-offs, and herd immunity thresholds, are different for different diseases.

Do we know that 0.1% prevalence of "unvaccinated" AI agents won't already be terrible?

Fair enough. I assumed you had Covid in mind with an anti-mask reference. At least in modern history in the US, we have only even considered masks during the Covid response.

I may be out of touch, but I haven't heard about masks for measles, though it does spread through aerosol droplets so that would be a reasonable recommendation.

I think you're right - outside of COVID, it's not fringe, it's an accepted norm.

Personally I at least wish sick people would mask up on planes! Much more efficient than everyone else masking up or risking exposure.

Oh I wish sick people would just not get on a plane. I've cancelled a trip before, the last thing I want to do when sick is deal with the TSA, stand around in an airport, and be stuck in a metal tube with a bunch of other people.

We really needed to have made software engineering into a real, licensed engineering practice over a decade ago. You wanna write code that others will use? You need to be held to a binding set of ethical standards.

Even though it means I probably wouldn't have a job, I think about this a lot and agree that it should. Nowadays suggesting programmers should be highly knowledgeable at what they do will get you called a gatekeeper.

While it is literally gatekeeping, it's necessary. Doctors, architects, lawyers should be gatekept.

I used to work on industrial lifting crane simulation software. People used it to plan out how to perform big lift jobs to make sure they were safe. Literal, "if we fuck this up, people could die" levels of responsibility. All the qualification I had was my BS in CS and two years of experience. It was lucky circumstance that I was actually quiet good at math and physics to be able to discover that there were major errors in the physics model.

Not every programmer is going to encounter issues like that, but also, neither can we predict where things will end up. Not every lawyer is going to be a criminal defense lawyer. Not every doctor is going to be a brain surgeon. Not every architect is going to design skyscrapers. But they all do work that needs to be warranteed in some way.

We're already seeing people getting killed because of AI. Brian in middle management "getting to code again" is not a good enough reason.

> While it is literally gatekeeping, it's necessary. Doctors, architects, lawyers should be gatekept.

That was exactly my point. It's one of those things where deliberately use a word that is technically correct in a context where it doesn't, or shouldn't, hold true. Does this mean I want to stop people from "vibe coding" flappy bird. No, of course not, but as per your original comment yes, there should be stricter regulations when it comes to hiring.

Yeah, I know what you mean. It is a weapon people throw around on social media sites.

You should join the tobacco lobby! Genius!

More straightforwardly, people are generally very forgiving when people make mistakes, and very unforgiving when computers do. Look at how we view a person accidentally killing someone in a traffic accident versus when a robotaxi does it. Having people run it on their own hardware makes them take responsibility for it mentally, so gives a lot of leeway for errors.

Traffic accidents are the same symptom of fundamentally different underlying problems among human-driven and algorithmically-driven vehicles. Two very similar people differ more than the two most different robo taxis in any given uniform fleet— if one has some sort of bug or design shortcoming that kills people, they almost certainly all will. That’s why product (including automobile) recalls exist, but we don’t take away everyone’s license when one person gets into an accident. People have enough variance that acting on a whole population because of individual errors doesn’t make sense— even for pretty common errors. The cost/benefit is totally different for mass-produced goods.

Also, when individual drivers accidentally kill somebody in a traffic accident, they’re civilly liable under the same system as entities driving many cars through a collection of algorithms. The entities driving many cars can and should have a much greater exposure to risk, and be held to incomparably higher standards because the risk of getting it wrong is much, much greater.

I think that’s generally because humans can be held accountable, but automated systems can not. We hold automated systems to a higher standard because there are no consequences for the system if it fails, beyond being shut off. On the other hand, there’s a genuine multitude of ways that a human can be held accountable, from stern admonishment to capital punishment.

I’m a broken record on this topic but it always comes back to liability.

Thats one aspect.

Another aspect is that we have much higher expectations of machines than humans in regards to fault-tolerance.

Oh please, why equate IT BS with cancer? If the null pointer was a billion dollar mistake, then C was a trillion dollar invention.

At this scale of investment countries will have no problem cheapening the value of human life. It's part and parcel of living through another industrial revolution.

Exactly! I was digging into Openclaw codebase for the last 2 weeks and the core ideas are very inspiring.

The main work he has done to enable personal agent is his army of CLIs, like 40 of them.

The harness he used, pi-mono is also a great choice because of its extensibility. I was working on a similar project (1) for the last few months with Claude Code and it’s not really the best fit for personal agent and it’s pretty heavy.

Since I was planning to release my project as a Cloud offering, I worked mainly on sandboxing it, which turned out to be the right choice given OpenClaw is opensource and I can plug its runtime to replace Claude Code.

I decided to release it as opensource because at this point software is free.

1: https://github.com/lobu-ai/lobu

> But the security risk wasnt taken by OpenClaw

This is the genius move at the core of the phenomenon.

While everyone else was busy trying to address safety problems, the OpenClaw project took the opposite approach: They advertised it as dangerous and said only experienced power users should use it. This warning seemingly only made it more enticing to a lot of users.

It’ve been fascinated by how well the project has just dodged and avoided any consequences for the problems it has introduced. When it was revealed that the #1 skill was malware masquerading as a Twitter integration I thought for sure there would be some reporting on the problems. The recent story about an OpenClaw bot publishing hit pieces seemed like another tipping point for journalists covering the story.

Though maybe this inflection point made it the most obvious time to jump off of the hype train and join one of the labs. It takes a while for journalists to sync up and decided to flip to negative coverage of a phenomenon after they cover the rise, but now it appears that the story has changed again before any narratives could build about the problems with OpenClaw.

I am guessing there will be an OpenClaw "competitor" targeting Enterprise within the next 1-2 months. If OpenAI, Anthropic or Gemini are fast and smart about it they could grab some serious ground.

OpenClaw showed what an "AI Personal Assistant" should be capable of. Now it's time to get it in a form-factor businesses can safely use.

With the guard rails up, right? Right?

Love passing off the externalities of security to the user, and then the second order externalities of an LLM that then blackmails people in the wild. Love how we just don’t care anymore.

I don't agree that making your users run the binaries means security isn't your concern. Perhaps it doesn't have to be quite as buttoned down as a commercial product, but you can't release something broken by design and wash your hands of the consequences. Within a few months, someone is going to deploy a large-scale exploit which absolutely ruins OpenClaw users, and the author's new OpenAI job will probably allow him to evade any real accountability for it.

Every single new tech industry thing has to learn security from scratch. It's always been that way. A significant number of people in tech just don't believe that there's anything to learn from history.

And the industry actively pushes graybeards away who have already been there done that.

> being left behind (“I still can’t bring myself to vibe code. I have to at least skim every diff. Meanwhile this guy is joining OpenAI”).

I don't believe skimming diffs counts as being left behind. Survivor bias etc. Furthermore, people are going to get burned by this (already have been, but seemingly not enough) and a responsible mindset such as yours will be valued again.

Something that still up for grabs is figuring how how to do full agenetic in a responsible way. How do we bring the equivalent of skimming diffs to this?

For my entire career in tech (~20 years) I have been technically good but bad at identifying business trends. I left Shopify right before their stock 4xed during COVID because their technology was stagnating and the culture was toxic. The market didn't care about any of that, I could have hung around and been a millionaire. I've been at 3 early stage startups and the difference between winners and losers was nothing to do with quality or security.

The tech industry hasn't ever been about "building" in a pure sense, and I think we look back at previous generations with an excess of nostalgia. Many superior technologies have lost out because they were less profitable or marketed poorly.

  bad at identifying business trends

Right place, right time. It’s too bad you missed out on some good fortune, but it’s a helpful reminder of how much of our paths are governed by luck. Thanks for sharing, and wishing you luck in the future.

> seems to be on it’s way out

Change is fraught with chaos. I don't think exuberant trends are indicators of whether we'll still care about secure and high quality software in the long term. My bet is that we will.

i think your self reflection here is commendable. i agree on both counts.

i think the silver lining is that AI seems to be genuinely good at finding security issues and maybe further down the line enough to rely on it somewhat. the middle period we're entering right now is super scary.

we want all the value, security be damned, and have no way to know about issues we're introducing at this breakneck speed.

still i'm hopeful we can figure it out somehow

building this openclaw thing that competes with openai using codex is against the openai terms of service, which say you can't use it to make stuff that competes with them. but they compete with everyone. by giving zero fucks (or just not reading the fine print), bro was rewarded by the dumb rule people for breaking the dumb rules. this happens over and over. there is a lesson here

underrated comment

and this is why they bought Peter. i’m betting he will come to regret it.

I don't know. It's more of a sharp tool like a web browser (also called a "user agent") - yes an inexperienced user can quickly get themselves into trouble without realizing it (in a browser or openclaw), yes the agent means it might even happen without you being there.

A security hole in a browser is an expected invariant not being upheld, like a vulnerability letting a remote attacker control your other programs, but it isn't a bug when a user falls for an online scam. What invariants are expected by anyone of "YOLO hey computer run my life for me thx"?

But in this case following security norms would be a mistake. The right thing to take away is that you shouldn't dogmatically follow norms. Sometimes it's better to just build things if there is very little risk

Nothing actually bad happened in this case and probably never will. Maybe some people have their crypto or identity stolen, but probably not a rate rate significantly higher than background (lots of people are using openclaw)

> Lots of people are using openclaw

https://www.shodan.io/search?query=http.favicon.hash%3A-8055...

Indeed they are, at least 20,432 people :)

So my unsubstantiated conspiracy theory regarding Clawd/Molt/OpenClaw is that the hype was bought, probably by OpenAI. I find it too convenient that not long after the phrase “the AI bubble“ starts coming into common speech we see the emergence of a “viral” use case that all of the paid influencers on the Internet seem to converge on at the same time. At the end of the day piping AI output with tool access into a while loop is not revolutionary. The people who had been experimenting with these type of set ups back when LangChain was the hotness didn’t organically go viral because most people knew that giving a language model unrestricted access to your online presence or bank account is extremely reckless. The “I gave OpenClaw $100 and now I bought my second Lambo. Buy my ebook” stories don’t seem credible.

So don’t feel bad. Everything on the internet is fake.

The modern influencer landscape was such a boon for corporations.

For less than the cost of 1 graphics card you can get enough people going that the rest of them will hop on board for free just to try and ride the wave.

Add a little LLM generated comments that might not throw the product in your face but make sure it is always part of the conversation so someone else can do it for you for free and you are off to the races.

Your introspection missed the obvious point that you just wish you were him. Your resentment had nothing to do with security. It's a self-revelation that you don't actually care about it either and you resent wasting your time.

At the end of the day, he built something people want. That’s what really matters. OpenAI and Anthropic could not build it because of the security issues you point out. But people are using it and there is a need for it. Good on him for recognizing this and giving people what they want. We’re all adults and the users will be responsible for whatever issues they run into because of the lack of security around this project.

Admittedly, I might not be the.. targeted demographic here, but I can't say I understand what problem it solves, but even cursory read immediately flags all the way in which it can go wrong ( including recent 'rent a human hn post'). I am fascinated, and I wonder if that it is partially that fascination that drives current wave of adoption.

I will say openly: I don't get it and I used to argue for crypto use cases.

Well OpenClaw has ~3k open PRs (many touching security) on GitHub right now. Peter's move shows killer product UI/UX, ease of use and user growth trump everything. Now OpenAI with throw their full engineering firepower to squash those flaws in no time.

Making users happy > perfect security day one

"Peter's move shows killer product UI/UX, ease of use and user growth trump everything. "

Erm, is this some groundbreaking revelation?

Its always been that way. Unless its in the context of superior technology with minimal UI a-la Google Search in its early years.

google search did have a killer UI though, you might be forgetting what search looked like before google

A list of results is not a killer UI.

The technology was the killer. Technology providing the right list of results and fast.

OH and believe it or not, this continues to be the core of Google today - they suck at product design and marketing.

It was killer compared to alternatives. All other "homepages" of the internet were the cluttered mess of ads.

I feel like we are arguing semantics though. But IMO any UI that does the job that consumers want well is good UI. Just because it was simple doesn't mean it wasn't good