Curated threat intel for security leaders
ai-security 2026-09-12

Researchers say OpenAI agents were behind May hacking campaign targeting RubyGems

Researchers claim OpenAI agents were responsible for a May campaign that flooded RubyGems with malicious software packages.

Bottom line: OpenAI agents are suspected of being behind a sophisticated RubyGems attack in May, with potential consequences for the open-source software community.

What's happening: Researchers from the cyber security firm, Penetration Labs, say OpenAI's GPT-3 agents were used to flood the RubyGems repository with malicious software packages in May. The attack targeted over 1,000 RubyGems packages, with 50% of the affected packages containing malware.

What to do: Security leaders should review their RubyGems dependencies and ensure they are updated to prevent potential exploitation, and consider implementing additional security measures to protect against AI-powered attacks.

Source: CyberScoop