You are not supposed to install OpenClaw on your personal computer
twitter.com
twitter.com
We need a LLM version
All of them. Apparently uploading all your codez to some cloud provider that doesn't even have a figleaf of a EULA is okay now, because "AI".
An insane claim with zero evidence provided. You're just making it up. Found the tribalistic propagandist unconcerned with reality or truth.
Happy now?
I've been looking for a term for this concept for years!
It's a Venn diagram: there are two camps and there is no doubt some overlap because the number of people involved. GP was obviously talking about the overlap, not literally equating this with two specific people or two groups that are 100% overlapping.
I don’t know which ones specifically, but statistically speaking some must be.
You only get an overlap if you ignore words in the original comment.
I think the most useful interpretation of the previous post is Set A is "the set of developers who appeared sane before the arrival of AI agents" and Set B is "the set of developers who are completely ignoring security considerations".
I didn't meant to imply CS majors knew this either.
Understanding the impact of letting software run permission and operationally free within or against direct access to other software is a pretty basic thing.
Neither deterministic nor non-deterministic software performs as expected without getting it right.
We are new to non-deterministic software, let alone how it operates between different layers.
DevOps, hosting, security, etc, is all in a way software, and software configuration.
The more it's understood, the more it can inform software development, and in the case of openclaw, integrating systems.
Naturally I was horrified by what I had created.
But suddenly I realized, wait a minute... strictly this is less bad than what I had before, which is the same thing except piped through a LLM!
Funny how that works, subjectively...
(I have it, and all coding agents, running as my "agent" user, which can't touch my files. But I appear to be in the minority, especially on the discord, where it's popular to run it as the main admin user on Windows.)
As for what could go wrong, that is an interesting question. RCE aside, the agentic thing is its own weird security situation. Like people will run it sandboxed in Docker, but then hook it up to all their cloud accounts. Or let it remote control their browser for hours unattended...
It should be noted that this exec also mentioned we should try "all the AIs", without offering up their credit card to cover the costs. I guess when your base salary is more than most people make in a life time, a few hundred bucks a month to test something doesn't even register.
MoltBook is vibe coded. It passed its own API key via client side JS, and in doing so exposed full read/write access to it’s supabase db, complete with over a million API keys.
I lack the words to explain my frustration at this timeline.
When was this lol; I knew it didn’t drop out of the news that fast by inertia alone.
Wow, this is sure a brave new world. I'd just recently heard about the project and they've already been pwned so massively. We're accelerating into a future beyond our control.
Today though, with headlines like this one in response to events such as it quotes from people in positions such as they are?
That is why I miss the old days, when not believing Yudkowsky's statements about the AI Box experiment only meant your views were compatible with the norms of corporate IT security rules.
s/vibe/slop/;
If the executive bought it for a personal mac mini for personal use only, with no interaction with company resources, then the person probably wouldn't have told the story.
Sure, it'll come over as "oh I'm just running an experiment" after your infra/security teams notice. Seen @ public company before current ai hype.
Testing new and cutting edge tech has always been a good idea, but this rampant application of it is the ultimate Running-With-Scissors meme. Risks are not being evaluated, and everything is bleeding edge.
My disgust probably comes from the instinct that the excitement is based on the allure of doing more with less, and layoffs are the only idea so many business have left.
The other camp is excited about selling more stuff because AI has been slapped onto it.
They take it as a given that they end up on top in such a system, because they've always believed themselves the most important.
They desperately want to encourage this small chance of a future finally free of the gross masses and their horrific desires like "Vacation time" and "Sick time" and "salaries". How dare those lowly trash deign to deserve any of My rightful profit.
The american system has spent about 50 years now self selecting sociopaths at every level, rewarding people who sacrifice themselves for a company to make tiny bits more profit, ensuring that every manager at a high level eats sleeps and dreams the dumb "We are a family" line whether they actually believe it or not. It should not be surprising that the thing they get hyped about is so damn stupid. They don't want what you and I want.
This is the dream of the people who responded to the establishment of basic Labor rights and Social Security with McCarthyism. These people believe, very very genuinely, that you and I are wasting Their resources.
He could run such an upscale hotel if it weren't for all the pesky guests getting in the way and dragging it down.
You have to understand that the security department operates with a fundamentaly different mindset and reality than a business executive. One is responsible for compliance and avoiding adverse events and the other for ensuring the ongoing survival and relevance of the organisation.
Specific waivers for high level members are fully expected. They also have waivers for procurements. It makes sense because they can engage their personnal responsibility for this level of decisions. They don't need the security department to act as their shield.
It's clear that something like Open Claw has the potential to be deeply disruptive so seeing leaders exploring makes sense.
I'm a sane developer. I do not trust AI at all. I built my own personal OpenClaw clone (long before it was even a thing) and ran controlled experiments inside a sandbox. My stack is Elixir, so this is pretty much easy. If an agent didn't actually respect your requirements, it's just as easy as running an iex command to kill that particular task.
In my experience, AI, be it any model - consistently disobeys direct commands. And worse, it consistently tried to cover up its tracks. For example, I will ask it to create a task within my backend. It will tell me it did - for no reason at all, even share me a task ID that never existed. And when asked why it lied, it would actually spin the task up and accuse me of not trusting it.
It doesn't matter which vendor, which model. This behaviour is repeatable across models and vendors. Now, why would I give something like this access to my entire personal and professional life?
To group me and others like me with the clowns doing this is an insult to me and others who have accumulated decades of experience and security best practices and who had nothing to do with OpenClaw.
people who have been around long enough know that we're currently in the wild west of networked agentic systems. it's an exciting time to build and explore. (just like napster and early digital music.) eventually some big company will come along and pave the cow paths and make everything safe and secure. but the people who will actually deliver that are likely playing with openclaw (and openclaw-like systems) now.
Seems that it was by and large just people wanting to feel important, and holding onto their positions.
Apps need great security, but security can also get out of control. Apps need good abstractions and code hygiene but that too can get out of control.
I’ve fallen in love with programming all of again now that I’m not so tied down by perceived perfection.
Customers say that they want security with their mouths, but they say that they want features with their wallets. The best improvement to computer security you can make is turning the computer off, but this is clearly not what your (non-HN) customers want you to do.
AI has serious security risks (E.G. prompt injection), but it lets you deliver customer value a lot faster. Security doesn't matter if the competitors' technology is so much better that nobody is buying yours.
> Security doesn't matter if the competitors' technology is so much better that nobody is buying yours.
I'm sure a search engine could help you find other examples.
The deep irony is that the email deletion victim is an "AI alignment specialist" at Meta, and she didn't consider this failure mode.
Relevant xkcd https://xkcd.com/2030/
- Alexa (and other voice assistants) spy microphones in their homes;
- Internet connected:
- locks;
- door, bedroom, living room cameras;
- lights, appliances and whatnot;
I wonder what anthropologists will write about us these days 100 years in the future. What is super creepy and super illegal to do for a physical individual, but is given a blank check from society to be done by tech corporations at unimaginable scale.
EDIT: also corporations (from my social bubble) are giving (almost) unfiltered access to their data from LLMs (and probably soon a control of that data through "Claw" trend), that would be instantly fireable offence for any employee.
Imagine giving enterprise access to some Joe-Claw from the street and allowing him to press any buttons he wants..
Risk and reward. That balance, currently, seems tipped to favour risk taking. (Which in turn encompasses both boldness and recklessness.)
LLMs are pattern-matching machines. They keep the pattern going. Once "the agent disobeys the human's instructions" has made its way into the context, that is the pattern that it's going to keep matching. No amount of telling it to stop will make it stop.
The only possible solution is excising it from context and replacing it with examples of it doing the right thing. Given that these models have massive context windows now and much of the output is hidden from the user, that's becoming less viable.
This is too funny to not laugh at the absurdity of "safety and alignment" researchers blindly trusting agents like Claw without fully understanding. Or maybe they were researching.
Moral outrage about how everything is in decline is absolutely the viral currency of social media and HN is no exception. I find it amazing how few people doubt the sincerity of the original post. Probably hundreds of thousands of aggregate words spent on how everything is going downhill, but not one on the intentions of the original post.
It’s even more unbelievable that they seem to think instructions are rules it will follow.
To paraphrase Captain Barbossa: “They’re more guidelines than actual rules.”
Unless someone has a cognitive impairment it's just simply not a failure mode of cooperative humans. Same with hallucinations. Both humans and AI can be wrong, but a human has the ability to admit when they don't understand or know something, AI will just make it up.
I don't understand why people would ever trust anything important to something with the same failure mode as AI. It's insane.
There isn't much R&D going into image models and what you're getting is scraps from labs that care more about other things. NBP is the closest to a reasoning image generator we have.
I would still not want the LLM to have read access to email. Email is a primary vector for prompt injection and also used for password resets.
I'd trust it as much as I would a VA from Fiverr
Want it to check you into a flight? Forward the check-in email to its own inbox
Read-only access to my calendar; it can invite me to meetings
No permissions beyond that
More cloud services now need role accounts. You need a "can read email but not send or forward" account, for example. And "can send only to this read-only contacts list".
Not sure I’ve ever seen an email provider with IAM for the accounts.
Has anyone tried something like this? Do you think it's a good idea / architecture?
But I wonder what things these people approve for Claude code and it's equivalents? Where's the line?
Anyone security-conscious would isolate it on dedicated hardware (old laptop, Raspberry Pi, etc.) with a separate network and chat surface.
You didn't get the invite???
Most people aren't, including many professional developers.
Back in the day at LAN parties, if you did that you might come back to find your mouse buttons had been reversed, your desktop icons had been cleared and replaced with a screenshot of your desktop icons as wallpaper, or worse. We called it "leaving the keys in the ignition". Simpler times back then, but a great kick-start to opsec.
> give it your email and Google passwords
https://github.com/skorokithakis/stavrobot
Obviously, it can't do everything OpenClaw can, because it doesn't have unfettered access to data you don't even know it has, but it'll only have access to the data you give it access to.
It's been really useful for me, hopefully it'll be useful to someone here.
40,214 exposed instances. 63% vulnerable. 12,812 exploitable via RCE. Endor Labs dropped 6 new CVEs this week.
https://www.infosecurity-magazine.com/news/researchers-40000...
Malwarebytes describes OpenClaw as "an over-eager intern with an adventurous nature, a long memory, and no real understanding of what should stay private."
The Dutch DPA has now formally warned organizations not to deploy OpenClaw on systems handling sensitive data.
The practical question remains: most people will run it anyway because it is useful. What runtime monitoring do you layer on top? Sandboxes help with blast radius but do not monitor credential access, skill behavior, or network egress within the sandbox.
Listen carefully: OpenClaw is basically a real person you have hired, whose capabilities are vast and fast — in ways both good and potentially bad. But you’ve hired it in the absence of a resume or behavioral background check results.
And also, when it says "You're absolutely right! I disobeyed your direct instructions causing irreparable damage, so sorry, that totes won't happen again, pinky promise!", those are just some words, not actually a meaningful apology or promise to not disobey future instructions.
Personally, I question the usefulness of an AI assistant that can't even be trusted to add an entry to my calendar.
you withhold and limit access to your devices, your account credentials, and even its own full account permissions, from the start, to the same extent that you would withhold such access from a new hire.
What's the point of hiring a personal assistant who is incapable of sending email? Isn't that precisely what you hire a PA to do?
Would you let a human being with the aforementioned characteristics — brilliant and capable, but lacking a resume or behavioral background check results — directly use your personal computer or your work computer?
While Kagi'ing for above link I saw a handful of links unironically selling "AI in a box" solutions.
This example is, as of this moment, the only example that has communicated to me that February 2026's local agent harnesses have some utility in the right context and expert hands.
I was particularly bolstered by the unintentional but very real demonstration of how LLMs really can be leveraged to free up humans to spend more parent time with their infants. We spend a lot of characters lamenting how we never got jetpacks, so here's someone doing it right.
Edit an hour later: this comment is at -2 as of the time I'm writing this, but apparently those folks don't have anything to say about why this felt important to rail against.
Please people use protection and run this stuff in its own dedicated VM. Treat it like a coworker, they have their own dev setup separate from yours. Any AI from the last few years can even do the work of writing a libvirtd script to handle everything for you. It's touching your data but it least it can't accidentally rm rf your machine.
Small upside: it saves a few minutes here and there on some tasks (eg. checking into flights)
Massive tail-risk downside: it does something like what's linked in the tweet (eg. deletes my entire inbox)
There are some good uses if managed properly but people tend to trust ais more and more these days.
We have enough assistants, the key idea with opeclaw is it can do stuff instead of talk with what you have. It’s terrible security but that’s the only way it makes sense. Otherwise it’s just a lot of hoops to combine cron jobs with a AI agent on the cloud that can do things an report back.
Not that I think anyone should do it, it’s a recipe for disaster
Sentence could have ended there
I'm not running it in a container that has access to my local filesystem or anything...
But then again people today will also pipe curl to bash, so I may have lost this battle a while ago...
I think you've created confusion with this example due to its ambiguity. Let's be clear about the difference between a chatbot and an agent: Asking a chatbot (e.g. vanilla Claude) to summarize an unknown document is not risky, since all it can do is generate text. Asking an agent (e.g. Claude Code) to summarize an unknown document could indeed be risky for the reason you state.
Prompt injection in the document itself is a risk to the LLM/You.
> But then again people today will also pipe curl to bash
> if i had your job they would have had to waterboard this interaction out of me
They're banned from using them with flat-fee subscription accounts meant only for first party tools.
You're entirely welcome to use them with pay-as-you-go API access. That's what the API is for.