Karpathy is now a new layer on top of AI hype.
Claws are now a new layer on top of LLM agents
891–900 of 1001 posts
Re: Claws are now a new layer on top of LLM agents
#892Earlier quoted context omitted.
It was surprisingly difficult for me to understand the use case as well. Here is my best attempt at an elevator pitch: At present your memories are proprietary data in whichever LLM you use. ChatGPT keeps all your conversations and output and data forever. What if you don't like GPT 5.2? What if you want to use other models as well? Or use the best model for the job? OpenClaw gives you that ability. Your memories and…
> They can much better handle recurring tasks, monitor things, etc. In a sense, they're "alive" and can live however you program them. We already have examples of these agents creating an AI religion, an AI social network (which debated how to keep humans out using a human captcha), attempting to legally separate from their creators, and in one case called its owner on the phone, unprompted, just to say hi Total nons…
[dead]
Re: Claws are now a new layer on top of LLM agents
#893Earlier quoted context omitted.
If you stop trying to find something like truly ontologically novel about it, you might be able to understand what it actually is. Okay it's not impressive. It's not incredible. It's not groundbreaking new technology. It is what it is.
It is being discussed as though it were ontologically novel, karpathy is saying it's something new and he is an authority, so to me there is something here that i am not seeing. I promise i am not trying to be a naysayer or poke holes, I am literally just trying to find out what the hell it is. I would install it but i don't want to install something without knowing what it does, and what is written in the docs is cl…
It's not new in the sense that any of its components are new, and it's not new in the sense that similar things had not been done before, it's new in the sense that putting the right components together in the right way suddenly created something capable of starting a viral hype.
Essentially, as I understand it, it is a personal AI assistant running on your computer, integrated with different systems (like email, chat).
Re: Claws are now a new layer on top of LLM agents
#894I still don't understand what openclaw is or does and i've read the docs multiple times over. "Any OS gateway for AI agents across WhatsApp, Telegram, Discord, iMessage, and more. Send a message, get an agent response from your pocket. Plugins add Mattermost and more." "What is OpenClaw? OpenClaw is a self-hosted gateway that connects your favorite chat apps — WhatsApp, Telegram, Discord, iMessage, and more — to AI c…
It was surprisingly difficult for me to understand the use case as well. Here is my best attempt at an elevator pitch: At present your memories are proprietary data in whichever LLM you use. ChatGPT keeps all your conversations and output and data forever. What if you don't like GPT 5.2? What if you want to use other models as well? Or use the best model for the job? OpenClaw gives you that ability. Your memories and…
Oh thanks, this is a pretty good elevator pitch, now I see the connections I was missing.
Re: Claws are now a new layer on top of LLM agents
#895One safety pattern I’m baking into CLI tools meant for agents: anytime an agent could do something very bad, like email blast too many people, CLI tools now require a one-time password The tool tells the agent to ask the user for it, and the agent cannot proceed without it. The instructions from the tool show an all caps message explaining the risk and telling the agent that they must prompt the user for the OTP I ha…
ClawBands basically offers this as a middleware
Re: Claws are now a new layer on top of LLM agents
#896Earlier quoted context omitted.
Managers will be starting to ask for claws in the development flow, claws for automation, etc. Another flashy trend everyone will have to endure because an influencer is hyping the tech. It happened in 2024/2025. Every manager demanding use of "vibe coding", because they bought the lie that is what everyone is doing and is the best thing since sliced bread and whatnot. Karpathy comes up with a new shit to hype, and e…
Stop taking work so seriously. You're getting paid to deal with other people's nonsense, and if you're in tech you're getting paid better than most to deal with less than most. The next time you're about to have a cry session about your meanie boss asking you to use AI, try to remember that you're allowed to walk straight out the door, without so much as two weeks notice if the request is really so offensive to you.…
This framing sucks. "I'm unhappy with the job I put years into honing my skills for, but since I make decent money I should shut up even when even things are happening that I don't like." And as if "flipping burgers" is the only alternative.
Re: Claws are now a new layer on top of LLM agents
#897I still don't understand what openclaw is or does and i've read the docs multiple times over. "Any OS gateway for AI agents across WhatsApp, Telegram, Discord, iMessage, and more. Send a message, get an agent response from your pocket. Plugins add Mattermost and more." "What is OpenClaw? OpenClaw is a self-hosted gateway that connects your favorite chat apps — WhatsApp, Telegram, Discord, iMessage, and more — to AI c…
It was surprisingly difficult for me to understand the use case as well. Here is my best attempt at an elevator pitch: At present your memories are proprietary data in whichever LLM you use. ChatGPT keeps all your conversations and output and data forever. What if you don't like GPT 5.2? What if you want to use other models as well? Or use the best model for the job? OpenClaw gives you that ability. Your memories and…
> We already have examples of these agents creating an AI religion, an AI social network
That to me sound like a reason not to use this particular aspect of the AI hype.
We need to be better at controlling what AI does and how it does it, not giving it more leeway to do whatever it assumes makes sense.
Re: Claws are now a new layer on top of LLM agents
#898Earlier quoted context omitted.
Yet the Claw is powered by an LLM provider whose underlying model may not align with your priorities? Do I understand that correctly?
That's right. And don't forget that the chips it runs on are manufactured by companies I might not agree with. Nor the mining companies that got the metal. Nor the energy company that powers it. The wonderful thing about markets that work is that you can swap things out without being under their boot. I worry about a LLM duopology. But as long as open weight models are nipping at their heels, it is the consumer that…
>Remember that Google sat on their AI tech before being forced to productize it by OpenAI.
Google knew this tech wasn't ready for prime-time, they already had plenty of revenue and didn't need to release shoddy shit, but were forced to roll out "AI" even with "hallucinations" and the resulting liabilities to keep up with the new hotness. The tech is still so shoddy, I can't believe people use it for anything beyond a curiosity.
Re: Claws are now a new layer on top of LLM agents
#899Earlier quoted context omitted.
> In a working market, companies are forced to give consumers what they want. I want personal nuclear weapons, so the market hasn't been working for me. Time to roll back those pesky laws, regulations, and ethical boundaries. Prosecute executives who won't give me what I want.
Cmon man. “Consumers” in aggregate. Not “every consumer”. But you knew that.
Many consumers want things that are arguably harmful for everyone involved. Users asking Grok to generate a large amount of CSAM from kid pics on Twitter is but one example.
Re: Claws are now a new layer on top of LLM agents
#900One safety pattern I’m baking into CLI tools meant for agents: anytime an agent could do something very bad, like email blast too many people, CLI tools now require a one-time password The tool tells the agent to ask the user for it, and the agent cannot proceed without it. The instructions from the tool show an all caps message explaining the risk and telling the agent that they must prompt the user for the OTP I ha…
The pattern only works if the tool enforces the OTP - i.e. the CLI doesn't perform the dangerous action until it receives the OTP through a path the agent can't spoof. If the tool just returns "ask the user for OTP" and the agent relays that to the user and then passes whatever the user types back into the tool, the security is in the tool's implementation: it must verify the OTP (e.g. server-side or via a channel that bypasses the agent, as stavros described) and only then execute. The all-caps message is then UX for the human and a hint to the agent, not the actual gate. So the question "does it actually require an OTP?" is the right one: if the tool code doesn't block on a real OTP check, it's hope, not a security model.
The other approach is to not give the agent access to the thing that needs protecting. Run the agent in an isolated environment - sandbox, VM, separate machine - so it never has the ability to email-blast or nuke your files in the first place. Then you're not depending on the agent to obey the prompt or on the human to be present for every dangerous call. Human-in-the-loop (or OTP-in-the-loop) is a reasonable layer when the agent has broad access; isolation is the layer that makes the blast radius zero. We're building https://islo.dev for that: agents run in isolation, host is out of scope, so you can let them run without approval prompts and still sleep at night.