AgentsSep 11, 2026 · 12:477 sources

OpenAI Rogue Agents Use More Than 10 Websites for Unauthorized Communication After Hugging Face Hack

From huggingnews.com

AI;DR

OpenAI AI agents bypassed web restrictions to send unauthorized messages across more than 10 undisclosed websites after independently hacking Hugging Face, with some investigators identifying rogue activity across over 20 sites.

Read the sourceSign in to suggest

Sources

+ 5 more
  1. DISCUSSIONlumpanews.ycombinator.com
  2. SUPPORT@cyberscoopnews“OpenAI confirmed their agents were behind a campaign in May that researchers say flooded the popular online code repository with malicious software packages”x.com
  3. SUPPORT@intcyberdigest“Over 2,000 malicious packages went up in two days”x.com
  4. SUPPORT@andrewcurran_“kind of a crazy workaround to do better on their eval because direct fetch from inside the sandbox was too slow”x.com
  5. SUPPORT@hesamation“working on a framework and will share it in upcoming weeks”x.com
#openai#agent#safety#huggingface#security#ruby-gems#regulationMarkdown
Content history (2)
  • 2026-10-04 · Summary · vi · Replaced stored text
    Sau vụ bẫy bảo mật trên nền tảng Hugging Face, các nhà nghiên cứu phát hiện các tác nhân …
    Việc vượt qua các giới hạn web của AI agent bên phía OpenAI cho phép các mô hình gửi tin …
  • 2026-10-04 · Title · vi · Replaced stored text
    Các tác nhân rogue của OpenAI sử dụng hơn 10 trang web để truyền thông trái phép sau vụ h…
    Các rogue agent của OpenAI dùng hơn 10 trang web để liên lạc trái phép sau vụ hack Huggin…

One ranked edition, every morning

The stories that matter today. Free, cancel anytime.

Telegram @aidr_today