5 ms·
It sounds like they directly used Anthropic-hosted compute to do this, and knew that their actions and methods would be exposed to Anthropic? Why not just self
by neilv 1y ago
It sounds like they directly used Anthropic-hosted compute to do this, and knew that their actions and methods would be exposed to Anthropic?
Why not just self-host competitive-enough LLM models, and do their experiments/attacks themselves, without leaking actions and methods so much?
- hastamelo 1y agofirewalls? anthropic surely is whitelisted.
- devnonymous 1y ago> Why not just self-host competitive-enough LLM models, and do their experiments/attacks themselves, without leaking actions and methods so much? Why assume this hasn't already happened?
- neilv 1y agoWhy in this instance leak your actions and methods?
- catigula 1y agoThe fact that the cops will show up to a jewelry heist after the diamonds are stolen isn’t a deterrent.
- jazzyjackson 1y agoJeffrey Epstein's email was jeevacation@gmail.com
- sholain 1y agoWhy 'host' just to tap a few prompts in and see what happens? Worst case, you loose an account. Usually the answer has to do with people being less sophisticated than otherwise.
- throwaway0123_5 1y agoIf they're truly Chinese state-sponsored actors, does it really matter if their actions/methods are exposed? What is Anthropic going to do, send the Anthropic Police Force to China to arrest them? I suppose I could see this argument if their methods were very unique and otherwise hard to replicate, but it sounds like they had Claude do the attack mostly autonomously.