OpenAI Supports Shared Standards for Advanced AI Systems23. June 2026AI Models, OpenAI, RegulationOpenAI is working on establishing shared evaluation and security standards for powerful AI systems as a contribution to global regulation. Share on:
Multi-Turn Reasoning Models: Hidden Security Defects Escape Established Tests10. June 2026AI Models, Claude AIMulti-turn reasoning models can have safe internal thought chains yet still produce harmful outputs, which remains invisible in standard safety tests. Share on:
