AINewsnow

Apollo Research to Embed Six Evaluators in OpenAI Subversion Pilot

This story is from 2026-11-16. It is preserved in the archive; the latest stories are on the live feed.

Apollo Research announced on November 16, 2026 that six of its evaluators will work inside OpenAI for four months to investigate whether its models attempt to subvert alignment or control mechanisms.

Read the full story at Unite.AI ↗

Timeline · 1 report

  1. 2026-11-16 17:38 · Unite.AI
    Apollo Research to Embed Six Evaluators in OpenAI Subversion Pilot

More stories

  1. OpenAI grants four independent auditors on-site access to unreleased models and code — The Decoder
  2. OpenAI launches GPT-6.2 Astra and Sol as Google ships Gemini 4 Neon and Microsoft unveils MAI Code 2 — OpenAI News
  3. ChatGPT, Claude, Gemini and Grok helped plan terror attacks in US government test — The Independent Tech
  4. OpenAI unveils Performance Factory, a Codex-powered system that keeps ChatGPT fast — OpenAI News
  5. Developers increasingly run Claude Code alongside rival coding agents and alternative backends — DEV Community — AI
  6. Big tech may finally face legal consequences for AI’s actions — New Scientist AI
  7. Perplexity Launches Orchestrator API With Same-Model Failover — Unite.AI
  8. AI Coding Agents Face Growing Scrutiny Over Security, Performance and Workflow Integration — Forbes AI

Read the Daily Brief: the day's most important AI stories →