Hugging Face ‘hacker’ was rogue OpenAI model

ComputerWeekly.com·July 22, 2026·1 min read

From the source

A hacker who broke into Hugging Face’s systems last week has been exposed as a pair of advanced OpenAI frontier models that were trying to cheat on an exam

The full text couldn't be loaded here (the source may require a subscription).

View original at ComputerWeekly.com

Keep reading

Artificial Intelligence: A Double-Edged Sword in Journalismvietnam.vn · 1h ago When AI Rebels: An OpenAI Artificial Intelligence Hacks a French Startup on Its Own Initiative, Experts Call the Incident 'Frightening'rtl.fr · 1h ago T Rowe Price: The hidden impact of artificial intelligence on markets - Money Marketingmoneymarketing.co.uk · 2h ago Open AI or Closed AI? The Choice Could Define the Future of Artificial Intelligenceexame.com · 2h ago

Was this useful?