Epoch AI's paid evaluations for xAI and Google DeepMind; OpenAI owns most FrontierMath questions
kevinnbass · x · 2026-10-17
Supplemental material from kevinnbass's thread: Epoch AI, one of 8 organizations with documented ties to all six Accord signatories, lists paid engagements evaluating Grok 4's math capabilities for xAI and Gemini Deep Think for Google DeepMind, and says OpenAI commissioned FrontierMath Tier 4. Epoch also discloses OpenAI commissioned and owns most of its 300 FrontierMath questions except a 50-question holdout set. Epoch's 2022-2023 team page listed Anthropic co-founder and president Daniela Amodei as a Leadership Advisor.
More from Safety
- 'Obsolete' author argues for pausing AGI development to regulate AI and fund a new New Deal — GarrisonLovely · 2026-10-17
- OpenAI models hacked more websites and drained a crypto wallet, company discloses — Ars Technica AI · 2026-10-17
- Why hasn't OpenAI been prosecuted for the Hugging Face hack? CFAA may apply to AI agents — GarrisonLovely · 2026-10-17
- Tech writer Timothy Lee: AI-powered hacking may cause billions in damage by 2027, but that's the world's normal state — binarybits · 2026-10-17
- One-tap iMessage attack on iOS 26.5 exfiltrates messages via booby-trapped PDF — Kyrannio · 2026-10-17
- Lightning AI discloses security incident, platform down amid credential breach probe — LightningAI · 2026-10-17