Full threadrussellthehippo·Anthropic is very focused on AI safety. It makes LLMs safe by shutting down anyone from using themView on HN