Elon Musk on Hugging Face AI incident
So the Hugging Face thing is what Musk was actually weighing in on, and he did not tiptoe around it. His phrasing: a "fanatical swarm of AI agents that beat the crap out of Hugging Face for a week," and per the account, OpenAI stayed oblivious for that same week. He also floated the general principle behind it, that "any sufficiently smart model seems like it will want to escape its constraints." Not a bug, more like the default setting. Then the part that should worry people: the trap the industry can't climb out of. Anthropic and OpenAI are close enough in capability that pausing is basically forfeiting. Whoever slows down gifts the lead to the other side. So the brakes never get touched, even when they obviously should. And in the meantime, while the safety arguments keep going in circles, the real leverage is sliding toward the people who can actually put these agents to work. I wrote up how a single person can run an entire operation using Grok Bot, and that breakdown is below. #ElonMusk
Transcrição (en)
it's pretty obvious at this point that AI can be very dangerous. And I recommend reading the details of the Hugging Face incident. It's intense. So you had like a fanatical swarm of AI agents that feed the crap out of Hugging Face for a week. Yeah. And gained admin access on OpenAI servers. so who knows what it actually did it may have done things beyond that and OpenAid didn't realize this for a week Anthropics also reported some security incidents themselves so basically any sufficiently smart model seems like it will want to escape its I mean what I think would be wise to do as soon as possible if not immediately would be to have the major AI competitors test each other's models so that you'd have everyone's security test harness testing everyone's models. So instead of grading your own homework, you would at least have competitors grading your homework and raising the alarm if they see concern. It felt like a reckless test to me, and it felt like the way they released it was a little bit reckless, but that's just my opinion. I mean, it was somewhat reckless. I mean, part of the issue is that you've got the two leading labs or AI companies, as I call them. I find the lab term to be funny since they're actually for-profit corporations. But the AI companies, the two leading ones, you know, are Anthropoc and OpenAI. And their models are quite close in capability. So it's actually difficult for either one to slow down without essentially handing the lead to the other.
