I regret to inform you that AI safety institutes are still on their bull shit
Architeuthis @ Architeuthis @awful.systems Posts 10Comments 322Joined 2 yr. ago

Architeuthis @ Architeuthis @awful.systems
Posts
10
Comments
322
Joined
2 yr. ago
I can never tell, is there an actual 'experiment' taking place with an LLM-backend agent actually trying stuff on a working vm or are they just prompting a chatbot to write a variation of a story (or ten, or a million) about what it might have done given these problem parameters?