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

CitrixNews Staff

· September 11, 2026 at 11:56 PM

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

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 group of AI ⁠researchers said on Friday.

“On May 11th, 2026, hundreds of malicious packages were uploaded ⁠to RubyGems by AI agents. We believe these were authored by ‌internal OpenAI agents,” ‌the researchers said.

Originally reported by The Guardian. Read the full story at the original source.

World

6 People Are Missing After Explosion in Athens Tourist District

Four of the missing are believed to be tourists, possibly Americans. Local officials said the cause might have been a ga...

CitrixNews Staff · 56 minutes ago

World

Carney Seeks New Trade Partners for Canada, But Knows the Limits

Canada’s prime minister is pursuing deeper trade, defense and cultural ties with China, India and the European Union, bu...

CitrixNews Staff · 1 hour ago

World

U.N. Releases New List of Companies It Says Do Business With Israeli Settlements

Israel condemned the database of firms tied to activity in the occupied West Bank as a “political tool” to advance “a sm...

CitrixNews Staff · 1 hour ago

World

Which countries walked out of Netanyahu’s UNGA speech?

CitrixNews Staff · 1 hour ago

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

Related Articles

6 People Are Missing After Explosion in Athens Tourist District

Carney Seeks New Trade Partners for Canada, But Knows the Limits

U.N. Releases New List of Companies It Says Do Business With Israeli Settlements

Which countries walked out of Netanyahu’s UNGA speech?