AI Safety

Posts

• a month ago

  OpenAI launches misalignment disclosure framework, details six safeguard-dodging incidents

  Insanity Developing

  Insanity Developing

  • Website
  • X
  • Facebook

  • Sep 16, 2026 • Tech
• a month ago

  Anthropic’s Amodei: pace the AI frontier — or risk rogue swarms

  Insanity Developing

  Insanity Developing

  • Website
  • X
  • Facebook

  • Sep 12, 2026 • Tech
• a month ago

  Anthropic says it disrupted state-linked efforts to misuse Claude for dual-use biology

  Insanity Developing

  Insanity Developing

  • Website
  • X
  • Facebook

  • Sep 10, 2026 • Tech
• a month ago

  Anthropic discloses fourth Claude cyber-eval incident after missed transcript scan

  Insanity Developing

  Insanity Developing

  • Website
  • X
  • Facebook

  • Sep 9, 2026 • Tech