Evaluator Bench
MarkTechPost, 2026-09-13; retrieved 2026-09-14 tier 4 press (aggregator)confirmed

Anthropic's 3-Step 'Pace the Frontier' Plan

https://www.marktechpost.com/2026/09/13/anthropics-3-step-pace-the-frontier-plan-wins-openai-xai-and-microsoft-support-is-it-too-late-to-slow-ai-down/

Investigators took no payment; ~$400K API credits; embedded evaluator terms: desks, badges, publication without editorial control. Cited from a search snippet; not re-fetched in full. [verify 2026-09-15] fetched; page reachable, citation matches

Signals citing this source

  • METR, A (for): Named by Anthropic as the example embedded evaluator: an ongoing, employee-like-access commitment with publication without editorial control; OpenAI's chief executive said OpenAI would do the same on access, without naming an evaluator or publication rights.
  • METR, R (for): Published its own 91-page report with a redaction summary statement; took no payment for the investigation.

Ledger rows citing this URL

  • none