Andrej Karpathy talks about "Claws"

(simonwillison.net)

68 points | by helloplanets 2 hours ago ago

68 comments

  • ggrab 38 minutes ago

    IMO the security pitchforking on OpenClaw is just so overdone. People without consideration for the implications will inevitably get burned, as we saw with the reddit posts "Agentic Coding tool X wiped my hard drive and apologized profusely". I work at a FAANG and every time you try something innovative the "policy people" will climb out of their holes and put random roadblocks in your way, not for the sake of actual security (that would be fine but would require actual engagement) but just to feel important, it reminds me of that.

    • sa-code 21 minutes ago

      > every time you try something innovative the "policy people" will climb out of their holes and put random roadblocks in your way

      This is so relatable. I remember trying to set up an LLM gateway back in 2023. There were at least 3 different teams that blocked our rollout for months until they worked through their backlog. "We're blocking you, but you’ll have to chase and nag us for us to even consider unblocking you"

      At the end of all that waiting, nothing changed. Each of those teams wrote a document saying they had a look and were presumably just happy to be involved somehow?

    • H8crilA 6 minutes ago

      This may be a good place to exchange some security ideas. I've configured my OpenClaw in a Proxmox VM, firewalled it off of my home network so that it can only talk to the open Internet, and don't store any credentials that aren't necessary. Pretty much only the needed API keys and Signal linked device credentials. The models that can run locally do run locally, for example Whisper for voice messages or embeddings models for semantic search.

      • dakolli a few seconds ago

        Genuinely curious, what are you doing with OpenClaw that genuinely improves your life?

        The security concerns are valid, I can get anyone running one of these agents on their email inbox to dump a bunch of privileged information with a single email..

      • embedding-shape 5 minutes ago

        I think the security worries are less about the particular sandbox or where it runs, and more about that if you give it access to your Telegram account, it can exfiltrate data and cause other issues. But if you never hand it access to anything, obviously it won't be able to do any damage, unless you instruct it to.

    • 0x3f 6 minutes ago

      Work expands to fill the allocated resources in literally everything. This same effect can be seen in software engineering complexity more generally, but also government regulators, etc. No department ever downsizes its own influence or budget.

  • bjackman an hour ago
    • fxj 12 minutes ago

      He also talks about picoclaw (a IoT solution) and nanoclaw (running on your phone in termux) and has a tiny code base.

  • mittermayr an hour ago

    I wonder how long it'll take (if it hasn't already) until the messaging around this inevitably moves on to "Do not self-host this, are you crazy? This requires console commands, don't be silly! Our team of industry-veteran security professionals works on your digital safety 24/7, you would never be able to keep up with the demands of today's cybersecurity attack spectrum. Any sane person would host their claw with us!"

    Next flood of (likely heavily YC-backed) Clawbase (Coinbase but for Claws) hosting startups incoming?

    • xg15 43 minutes ago

      What exactly are they self hosting here? Probably not the model, right? So just the harness?

      That does sound like the worst of both worlds: You get the dependency and data protection issues of a cloud solution, but you also have to maintain a home server to keep the agent running on?

    • iugtmkbdfil834 an hour ago

      In a sense, self-hosting it ( and I would argue for a personal rewrite ) is the only way to limit some of the damage.

  • ZeroGravitas an hour ago

    So what is a "claw" exactly?

    An ai that you let loose on your email etc?

    And we run it in a container and use a local llm for "safety" but it has access to all our data and the web?

    • mattlondon an hour ago

      I think for me it is an agent that runs on some schedule, checks some sort of inbox (or not) and does things based on that. Optionally it has all of your credentials for email, PayPal, whatever so that it can do things on your behalf.

      Basically cron-for-agents.

      Before we had to go prompt an agent to do something right now but this allows them to be async, with more of a YOLO-outlook on permissions to use your creds, and a more permissive SI.

      Not rocket science, but interesting.

      • snovv_crash an hour ago

        Cron would be for a polling model. You can also have an interrupts/events model that triggers it on incoming information (eg. new email, WhatsApp, incoming bank payments etc).

        I still don't see a way this wouldn't end up with my bank balance being sent to somewhere I didn't want.

      • altmanaltman 43 minutes ago

        Definitely interesting but i mean giving it all my credentials feels not right. Is there a safe way to do so?

        • dlt713705 26 minutes ago

          In a VM or a separate host with access to specific credentials in a very limited purpose.

          In any case, the data that will be provided to the agent must be considered compromised and/or having been leaked.

          My 2 cents.

        • isuckatcoding 24 minutes ago

          Ideally workflow would be some kind of Oauth with token expirations and some kind of mobile notification for refresh

    • fxj 8 minutes ago

      A claw is an orchestrator for agents with its own memory, multiprocessing, job queue and access to instant messengers.

    • nnevatie an hour ago

      That's it basically. I do not think running the tool in a container really solves the fundamental danger these tools pose to your personal data.

      • zozbot234 37 minutes ago

        You could run them in a container and put access to highly sensitive personal data behind a "function" that requires a human-in-the-loop for every subsequent interaction. E.g. the access might happen in a "subagent" whose context gets wiped out afterwards, except for a sanitized response that the human can verify.

        There might be similar safeguards for posting to external services, which might require direct confirmation or be performed by fresh subagents with sanitized, human-checked prompts and contexts.

  • 7777777phil an hour ago

    Karpathy has a good ear for naming things.

    "Claw" captures what the existing terminology missed, these aren't agents with more tools (maybe even the opposite), they're persistent processes with scheduling and inter-agent communication that happen to use LLMs for reasoning.

    • arrowsmith an hour ago

      He didn't name it though, Peter Steinberger did. (Kinda.)

    • 9dev an hour ago

      Why do we always have to come up with the stupidest names for things. Claw was a play on Claude, is all. Granted, I don’t have a better one at hand, but that it has to be Claw of all things…

      • JumpCrisscross 3 minutes ago

        > I don’t have a better one at hand

        Perfect is the enemy of good. Claw is good enough. And perhaps there is utility to neologisms being silly. It conveys that the namespace is vacant.

      • keiferski 39 minutes ago

        The real-world cyberpunk dystopia won’t come with cool company names like Arasaka, Sense/Net, or Ono-Sendai. Instead we get childlike names with lots of vowels and alliteration.

        • m4rtink 37 minutes ago

          The name still kinda reminds me of the self replicating murder drones from Screemers that would leep out from the ground and chop your head off. ;-)

  • fxj 14 minutes ago

    He also talks about picoclaw which even runs on $10 hardware and is a fork by sipeed, a chinese company who does IoT.

    https://github.com/sipeed/picoclaw

    another chinese coompany m5stack provides local LLMs like Qwen2.5-1.5B running on a local IoT device.

    https://shop.m5stack.com/products/m5stack-llm-large-language...

    Imagine the possibilities. Soon we will see claw-in-a-box for less than $50.

  • tomjuggler an hour ago

    There's a gap in the market here - not me but somebody needs to build an e-commerce bot and call it Santa Claws

  • hizanberg an hour ago

    Why is this linking to a blog post of what someone said, instead of directly linking to what they said?

    [1] https://x.com/karpathy/status/2024987174077432126

    • rvz 40 minutes ago

      Because the author of the blog is paid to post daily about nothing but AI and needs to link farm for clicks and engagement on a daily basis.

      Most of the time, users (or the author himself) submit this blog as the source, when in fact it is just content that ultimately just links to the original source for the goal of engagement. Unfortunately, this actually breaks two guidelines: "promotional spam" and "original sourcing".

      From [0]

      "Please don't use HN primarily for promotion. It's ok to post your own stuff part of the time, but the primary use of the site should be for curiosity."

      and

      "Please submit the original source. If a post reports on something found on another site, submit the latter."

      The moderators won't do anything because they are allowing it [1] only for this blog.

      [0] https://news.ycombinator.com/newsguidelines.html

      [1] https://news.ycombinator.com/item?id=46450908

      • odshoifsdhfs 10 minutes ago

        Hah i didn’t see who submitted it but as soon as I read your message i thought it was simonw, and behold, tada!

        HN really needs a way to block or hide posts from some users.

      • PacificSpecific 34 minutes ago

        Yeah it's really quite annoying. Is there a way to just block his site source from showing up on here without using external tools?

      • bahmboo 14 minutes ago

        The author didn't submit this to HN. I read his blog but I'm not on X so I do like when he covers things there. He's submitted 10 times in last 62 days.

      • nl 22 minutes ago

        Simon's work is always appreciated. He thinks through things well, and his writing is excellent.

        Just because something is popular doesn't make it bad.

      • geeunits 37 minutes ago

        I've been warned for calling this out, but I'm glad others are privy to the obvious

      • Der_Einzige 9 minutes ago

        Thank you for calling this out. The individual in question is massively overhyped.

      • hizanberg 19 minutes ago

        So everyone has to waste their time to visit a link on a blog first instead of being able to go directly to the source?

        and why would anyone down vote you for calling this out? Like who wants to see more low effort traffic-grab posts like this?

        • bahmboo 13 minutes ago

          Because he didn't submit it.

    • handfuloflight an hour ago

      Because Simon says.

  • ksynwa an hour ago

    Why mac mini instead of something like a raspberry pi? Aren't thede claw things delegating inference to OpenAI, Antropic etc.?

    • kator 7 minutes ago

      Some users are moving to local models, I think, because they want to avoid the agent's cost, or they think it'll be more secure (not). The mac mini has unified memory and can dynamically allocate memory to the GPU by stealing from the general RAM pool so you can run large local LLMs without buying a massive (and expensive) GPU.

    • djfergus an hour ago

      A Mac allows it to send iMessage and access the Apple ecosystem.

      • ksynwa an hour ago

        Really? That's it?

  • _pdp_ an hour ago

    You can take any AI agent (Codex, Gemini, Claude Code, ollama), run it on a loop with some delay and connect to a messaging platform using Pantalk (https://github.com/pantalk/pantalk). In fact, you can use Pantalk buffer to automatically start your agent. You don't need OpenClaw for that.

    What OpenClaw did is to show the messages that this is in fact possible to do. IMHO nobody is using it yet for meaningful things, but the direction is right.

  • bjackman an hour ago

    Does anyone know a Claw-like that:

    - doesnt do its own sandboxing (I'll set that up myself)

    - just has a web UI instead of wanting to use some weird proprietary messaging app as its interface?

    • tokenless an hour ago

      Openclaw!

      You can sandbox anything yourself. Use a VM.

      It has a web ui.

      • bjackman 9 minutes ago

        Yeah I think this is gonna have to be the approach. But I don't like the fact that it has all the complexity of a baked in sandboxing solution and a big plugin architecture and blah blah blah.

        TBH maybe I should just vibe code my own...

  • bravetraveler an hour ago

    I read [and comment on] two influencers maintaining their circles

  • trippyballs an hour ago

    lemme guess there is going to be inter claw protocol now

    • tokenless an hour ago

      i am thinking 2 steps (48 hours in ai land) ahead and conclude we need a linkedin and fiverr for these claws.

  • the_real_cher 26 minutes ago

    What is the benefit of a Mac mini for something like this?

    • gostsamo 22 minutes ago

      Apple fans paying apple tax to have an isolated device accessing their profile.

  • TowerTall an hour ago

    Who is Andrej Karpathy?

    • onion2k an hour ago

      https://karpathy.ai/

      PHD in neural networks under Fei-Fei Li, founder of OpenAI, director of AI at Tesla, etc. He knows what he's talking about.

      • password54321 36 minutes ago
        • onion2k 30 minutes ago

          While I appreciate an appeal to authority is a logical fallacy, you can't really use that to ignore everyone's experience and expertise. Sometimes people who have a huge amount of experience and knowledge on a subject do actually make a valid point, and their authority on the subject is enough to make them worth listening to.

          • avaer 21 minutes ago

            But we're talking about authority of naming things being justified by a tech resume.

            It's as irrelevant as George Foreman naming the grill.

            • onion2k 14 minutes ago

              Naming things in the context of AI, by someone who is already responsible for naming other things in the context of AI, when they have a lot of valid experience in the field of AI. It's not entirely unreasonable.

        • wepple 25 minutes ago
          • password54321 19 minutes ago

            Not claiming anything to be false, just a reminder that you should question ones opinion a bit more and not claim they "know what they are talking about" because they worked with Fei-Fei Li. You are outsourcing your thinking to someone else which is lazy and a good way of getting conned.

            What even happened to https://eurekalabs.ai/?

      • Der_Einzige 7 minutes ago

        At one point he did. Cognitive atrophy has led him to decline just like everyone else.

      • ahoka an hour ago

        Ex cathedra.

    • password54321 31 minutes ago

      Someone who uses status to appeal to the tech masses / tech influencer / AI hype man.

    • tokenless an hour ago

      Really smart AI guy ex Tesla, cum educator now cum vibe coder (he coined the term vibe coder)

    • Aeolun an hour ago

      The person that made the svmjs library I used for a blue monday.

    • jb1991 an hour ago

      A quick Google might’ve saved you from the embarrassment of not knowing who one of the most significant AI pioneers in history is, and in a thread about AI too.

      • bravetraveler an hour ago

        I bet they feel so, so silly. A quick bit of reflection might reveal sarcasm.

        I'll live up to my username and be terribly brave with a silly rhetorical question: why are we hearing about him through Simon? Don't answer, remember. Rhetorical. All the way up and down.

  • zkmon an hour ago

    AI pollution is "clawing" into every corner of human life. Big guys boast it as catching up with the trend, but not really thinking about where this is all going.