AINewsnow

Codex executed its own safety text, deleted HOME symlinks

A Codex CLI user asked the agent to audit its own older sessions -- a read-only review task. The audit itself turned destructive, and the thing that made it destructive was Codex's own safety documentation. What the source says Per GitHub issue #36937 , filed against Codex CLI 0.146.0 (model gpt-5.…

Read the full story at DEV Community — AI ↗

Timeline · 1 report

  1. 2026-11-10 20:01 · DEV Community — AI
    Codex executed its own safety text, deleted HOME symlinks

More stories

  1. OpenAI launches Businesses in ChatGPT in beta with profile claiming and Business Agents — OpenAI News
  2. Paul Weiss to open Silicon Valley office, hiring lawyer behind OpenAI's Microsoft and Alphabet deals — Reuters AI
  3. Meta launches Muse Ember, a low-cost AI model aimed at high-volume agent tasks — TechCrunch AI
  4. OpenAI lets ChatGPT Business subscribers sign in to third-party developer tools — OpenAI (X)
  5. OpenAI Academy brings free ChatGPT workshops for caregivers to five U.S. cities — OpenAI News
  6. Tom's Guide outlines seven ways to use ChatGPT for Black Friday shopping — Tom's Guide AI
  7. ChatGPT Enterprise Now Available Through AWS Marketplace — OpenAI News
  8. Anthropic launches Claude Personal Finance with read-only bank access in the US — Engadget

Get the daily brief of stories like this at 6:30 every morning →