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

Reuters

Published Sep 12, 2026

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

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.

TechnologyAmazon raises starting pay to $20 an hour and adds Whole Foods discount

Wage increase brings Amazon’s US minimum pay in line with Costco and includes new grocery discount benefit

Michael Sainato

Technology‘Godfather of AI’ says tech regulation is nearing Covid-style pivot moment

Safety crisis makes it more likely that governments will be spurred into action, says Yoshua Bengio

Dan Milmo Global technology editor

TechnologyAmazon to launch Ring ‘neighbourhood watch’ app in UK amid US privacy fears

Doorbell firm aims to replace WhatsApp or Facebook groups with platform where people can post about safety or lost pets

Samuel Gibbs Consumer technology editor

The EIA Briefing

Keep reading, every morning.

The stories that matter, chosen by our editors and in your inbox before 7 a.m. — every weekday.

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

More on This Story

Keep reading, every morning.