Anthropic staff say AI could kill everyone. But its safety rule keeps getting weaker Oct 31, 2026 · Anthony Lopopolo Anthropic now counts on filters that block weapons requests. Until rivals slow down, the company has to decide if a model is too dangerous to build Read more »