OpenAI Brings in Apollo, METR, Transluce and Goodfire for Embedded Safety Assessments
OpenAI · x · 2026-11-17
OpenAI announced partnerships with Apollo Research, METR, Transluce AI, and Goodfire AI for "embedded assessments"—giving outside experts deep access to its internal models to independently scrutinize alignment and safeguards.
- Apollo Research: will test whether models could covertly act against their developers' intentions and whether safeguards would catch it, examining training, safety testing, and internal-use evidence.
- METR: will investigate misaligned behavior in OpenAI's largest RL run to date, including failures existing monitors may have missed; a broader monitor assessment is preliminarily agreed.
- Transluce AI: will audit post-training alignment evals for quality and coverage, build additional tests, and expand the pilot in 2027.
- Goodfire AI: will embed interpretability researchers to analyze model internals for training-run monitoring, debugging, and alignment.
Related event: OpenAI partners with four external groups for embedded safety assessments(2 posts)→
More from Models
- Model-picker screenshots surface unannounced GPT-6.1, Gemini 4 Hydrogen and Claude Opus 5.5 names — TomLikesRobots · 2026-11-17
- New MKA Benchmark: 200 Expert-Level Multimodal Tasks Across 83 Occupations — osanseviero · 2026-11-17
- MKA Benchmark: Gemini 4 Carbon Scores 54.3% at $5.63 Per Task vs Claude's $35.28 — _philschmid · 2026-11-17
- Gemini Decisions API Judges 1,000 Questions Over 256K Tokens With Zero Output Token Billing — _philschmid · 2026-11-17
- RWKV-8 RPU preview hits 12,500+ tokens/s real-time inference — kastnerkyle · 2026-11-17
- Gemini 4 Helix spotted in Google Antigravity, hinting at another Gemini 4 variant — feblet · 2026-11-16