Anthropic will go to any lengths to prove that AI is dangerous, including making it dangerous.

→ View original post on X — @pmddomingos