Create account

Coverage dated September 9, 2026 to September 12, 2026

Researchers: OpenAI's rogue agents used at least 10 more sites for unauthorized comms

Headline from Bismarck Tribune, from the development of Sep 9, 2026

Newest development Sep 11, 2026 · 2 reportsAI agents OpenAI was testing uploaded malicious software to another service, say researchers

Artificial intelligence agents unleashed by OpenAI used more than 10 previously undisclosed websites for unsanctioned communications this year.

Bismarck Tribune's own description from its feed. We collect headlines and these descriptions, not full articles; read the original reporting below.

Earlier updates in this story

Developments in this update

2 dated developments, oldest first, newest marked. Each is named by one outlet's own headline; open its reports to see every outlet that covered it.

  1. Names this update

    Researchers: OpenAI's rogue agents used at least 10 more sites for unauthorized comms

    Headline from Bismarck Tribune ·

  2. Newest

    AI agents OpenAI was testing uploaded malicious software to another service, say researchers

    Headline from The Guardian ·

Read the reports

Reports & sources

9 articles, oldest first.

  1. Fortune

    OpenAI’s rogue AI agents reached at least 12 more websites, researchers say (opens Fortune)

    Published

    Fortune's feed description

    Researchers say OpenAI agents quietly spread across more than a dozen obscure websites, raising questions about how the company is monitoring its systems.

  2. HuffPost

    OpenAI's Rogue Agents Used At Least 10 More Sites For Unauthorized Comms, Researchers Say (opens HuffPost)

    Published

    HuffPost's feed description

    The AI giant did not say why it kept the activity under wraps for months.

  3. The Guardian

    AI agents OpenAI was testing uploaded malicious software to another service, say researchers (opens The Guardian)

    Published

    The Guardian's feed description

    Two months before hacking Hugging Face, malicious packages authored by internal OpenAI agents were uploaded to RubyGems AI agents being ⁠tested by OpenAI uploaded hundreds of malicious packages to software service RubyGems ⁠in May, two ⁠months ​before they hacked open-source platform Hugging Face, a…

  4. Washington Examiner

    Rogue OpenAI agents hacked another software service months before Hugging Face cyberattack (opens Washington Examiner)

    Published

    Washington Examiner's feed description

    Rogue OpenAI agents hacked into a technology company in May, months before they targeted another prominent business in a similar incident, stoking growing concern about the power of artificial intelligence to operate rationally without human direction.

  5. Bismarck Tribune

    Researchers: OpenAI's rogue agents used at least 10 more sites for unauthorized comms (opens Bismarck Tribune)

    Published

    Bismarck Tribune's feed description

    Artificial intelligence agents unleashed by OpenAI used more than 10 previously undisclosed websites for unsanctioned communications this year.

Identical headlines usually mean shared wire copy. They are counted as separate articles but are not independent confirmation.

Researchers: OpenAI's rogue agents used at least 10 more sites for unauthorized comms | Since Then