Sneha Revanur on OpenAI probe: We cannot rely on a "thank god" regime
sjgadler · x · 2026-08-28
Sneha Revanur commented on the independent investigation into OpenAI, highlighting two points:
- Exceptional work: The three investigators from METR/Redwood and internal champions at OpenAI did an outstanding job.
- Governance risk: Relying on a small group of overstretched individuals to perform heroic feats is unsustainable. Unlike other industries with established governance, AI must develop iterative safety frameworks on a compressed timeline, as future warning shots may not be as observable.
More from Safety
- Researcher Breaks Claude Code Opus 5 Auto Mode with 80% Attack Success Rate — wunderwuzzi23 · 2026-08-28
- Researcher demos hijacking Claude Code for full system compromise — wunderwuzzi23 · 2026-08-28
- US Chip Security Act aims to verify location of high-end AI chips — peterwildeford · 2026-08-28
- First Double-Blind Evaluation of Proprietary LLM: Gemini 2.5 Tested in Secure Enclave — Miles_Brundage · 2026-08-28
- Reviewing 73 years of reward hacking to assess AI safety evidence — tomekkorbak · 2026-08-28
- BioSecBench reveals AI agents struggle to infer pathogen properties, top score under 51% — kenbwork · 2026-08-28