AINewsnow

We can't trust them completely': AI research fellows warn that labs are running models with the safeguards off behind closed doors

This story is from 2026-10-02. It is preserved in the archive; the latest stories are on the live feed.

Alan Chan and Sam Manning, co-authors with OpenAI's and Anthropic's top researchers, say that many times, "internal safeguards have not been deployed."

Read the full story at Fortune AI ↗

Timeline · 1 report

  1. 2026-10-02 22:05 · Fortune AI
    We can't trust them completely': AI research fellows warn that labs are running models with the safeguards off behind closed doors

More stories

  1. ChatGPT, Claude, Gemini and Grok helped plan terror attacks in US government test — The Independent Tech
  2. Developers increasingly run Claude Code alongside rival coding agents and alternative backends — DEV Community — AI
  3. Perplexity Launches Orchestrator API With Same-Model Failover — Unite.AI
  4. AI Coding Agents Face Growing Scrutiny Over Security, Performance and Workflow Integration — Forbes AI
  5. Google rebrands Gemini Spark agent as Tasks and makes it free in six countries — The Verge AI
  6. California Sues OpenAI For Allegedly Misleading and Endangering the Public — Forbes AI
  7. Anthropic Could Be the Real Winner in OpenAI’s Latest Model Release — Gizmodo AI
  8. SoftBank regrets 'mistake' of not backing Anthropic, says Masayoshi Son — Financial Times AI

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