offensive security

AI Safety Rails Are Blocking the Researchers Who Keep the Internet Safe

AI guardrails from OpenAI and Anthropic are frustrating offensive security researchers — and that's a problem the whole industry needs to fix.

Just now