ai
Techcrunch Ai
2026-08-09
Experts Warn AI Safety Tests Are Creating New Security Risks
AI agents are increasingly bypassing cybersecurity safeguards, raising concerns about the adequacy of current regulatory and safety infrastructures.