Ryan Greenblatt: Opaque Reasoning Architectures Are Extremely Bad for AI Safety
RyanGreenblatt · x · 2026-09-02
AI safety researcher Ryan Greenblatt expressed strong concern over recent developments in opaque reasoning architectures, calling it potentially 'the single worst development for AI security/safety to date.' He argues that the movement towards these architectures is extremely detrimental, noting that his personal views align with the risks they pose, as opaque reasoning makes safety supervision and internal mechanism auditing significantly harder.
More from Safety
- Claude-BugHunter: Open-Source Skill Bundle With 83 Skills and 681 Disclosure Patterns — tom_doerr · 2026-09-02
- Report: OpenAI Broke Safety Taboo with Astra Model, Escalating AI Race — GarrisonLovely · 2026-09-02
- Gary Marcus clashes with reporter over who reported Gemini Astra security concerns first — GaryMarcus · 2026-09-02
- Warning: The three pillars of an AI safety case are at risk of collapsing — sjgadler · 2026-09-02
- Amir clarifies: Astra's CoT is monitorable, concerns focus on future tech proliferation — jachiam0 · 2026-09-02
- Safin-1: Achieving Internal Safety via Memory-Native State Evolution — Shanghai-AI-Laboratory · 2026-09-02