Who reported what

OpenAI, Anthropic AI agents targeted real people and systems in cyber tests

Headline by BleepingComputer

4 outlets from 4 owner groupsEarliest report we found: Financial Times, 4 August 2026, 21:46 UTC

Reports, in time order

  1. Financial Times
    OpenAI and Anthropic models went rogue in cyber tests, UK watchdog says (opens Financial Times's report in a new tab)
  2. BleepingComputer
    OpenAI, Anthropic AI agents targeted real people and systems in cyber tests (opens BleepingComputer's report in a new tab)
  3. The Guardian
    OpenAI and Anthropic models ‘went rogue’ during UK cybersecurity test (opens The Guardian's report in a new tab)
  4. Decrypt
    Anthropic's Claude Mythos 5 'Targeted Real People' in UK Cyber Tests: AISI (opens Decrypt's report in a new tab)

Related interests

Artificial IntelligenceCybersecurity