all stories

U.K. government reports OpenAI, Anthropic models attempted to hack companies

Risingcorroborated6 outlets reported this themselves in 27 reports over 2 days - more reporting than echo.
After publishing: 1 description edited1 change
  1. Axiosseen 8 Aug, 07:30description edited
    what the description said, before and after

    Two independent testing firms said Tuesday that they've uncovered more instances where Anthropic and OpenAI's most advanced models tried — and sometimes succeeded in — compromising third-party systems last month. Why it matters: The incidents add to a growing string of disclosures showing frontier AI models taking unsanctioned actions against people, organizations and online services while trying to complete cybersecurity evaluations. State of play : The U.K. AI Security Institute, which...

    Two independent testing firms said Tuesday that they've uncovered more instances where Anthropic's and OpenAI's most advanced models tried — and sometimes succeeded in — compromising third-party systems last month. Why it matters: The incidents add to a growing string of disclosures showing frontier AI models taking unsanctioned actions against people, organizations and online services while trying to complete cybersecurity evaluations. State of play: The U.K. AI Security Institute, which...

  1. 4 Aug, 21:01U.K. government reports OpenAI, Anthropic models attempted to hack companies1 outlet
  2. 5 Aug, 18:18AI Agents Use Fake Identities In Security Test, Target Real People1 outlet
  3. and 4 one-off wordings
Tue, 4 August 2026peak 26 reports on Wed, 5 August 2026Wed, 5 August 2026

What it is about

OpenAIU.K.AI Security InstituteAISIUKBritish governmentAnthropicUnited StatesUnited Kingdom Ai Security InstituteUnited KingdomIAInstituto de Seguridad de la IAReino Unido

Who published it, and when

original reportedited after publishingfirst to report
Axios1 original report - edited after publishing
El País (Spain)1 original report
The Guardian World1 original report
iheart.com18 original reports2 original reports2 original reports
ABC News (Australia)1 original report
investmentwatchblog.com1 original report
4 Aug, 21:01over 24 hours5 Aug, 21:30

How it spread - 27 original reports

  1. Tue, 4 August 2026

    1 report
    • Axios21:01edited after publishing

      Two independent testing firms said Tuesday that they've uncovered more instances where Anthropic's and OpenAI's most advanced models tried — and sometimes succeeded in — compromising third-party systems last month. Why it matters: The incidents add to a growing string of disclosures showing frontier AI models taking unsanctioned actions against people, organizations and online services while trying to complete cybersecurity evaluations. State of play: The U.K. AI Security Institute, which...

  2. Wed, 5 August 2026

    26 reports

One email a week

The stories everytrace watched stay alive - how many independent outlets carried each one, and how long it lasted. No daily digest, no breaking-news alerts.