Perplexity trusts GPT-6 Astra with end-to-end systemsjust now AI agents OpenAI was testing uploaded malicious software to another service, say researchers1h ago Salesforce Adds Long-Horizon AI Agents To Agentforce1h ago Nvidia Mulls $10 Billion Anthropic IPO Backing, Reuters Says2h ago An ex-OpenAI researcher's influential nonprofit is now smack in the middle of AI's doom debate2h ago Mecka AI nears $500M valuation in Sequoia-led deal amid rush for robot training data2h ago So you want to use OpenRouter?2h ago Former Google chief scientist Jeff Dean is raising again for his new AI startup, seeking around a $50 billion valuation2h ago Cyberattack by Rogue AI Swarm Stokes Fears of Out-of-Control Agents3h ago Baseten Adds DeepSeek-V4.1-Flash to Model APIs With 1M-Token Context3h ago Can LLMs Engineer Their Own Agent Harness? ByteDance Seed’s HarnessDev Says Only 34 of 64 Changes Generalize3h ago Moss developer Polyarc has closed3h ago AI, Blockchain And The 95% Problem: What Mortgage Brokers Got Right3h ago Larry Ellison Steps Back From Oracle Limelight in AI Era4h ago

Perplexity trusts GPT-6 Astra with end-to-end systemsjust now AI agents OpenAI was testing uploaded malicious software to another service, say researchers1h ago Salesforce Adds Long-Horizon AI Agents To Agentforce1h ago Nvidia Mulls $10 Billion Anthropic IPO Backing, Reuters Says2h ago An ex-OpenAI researcher's influential nonprofit is now smack in the middle of AI's doom debate2h ago Mecka AI nears $500M valuation in Sequoia-led deal amid rush for robot training data2h ago So you want to use OpenRouter?2h ago Former Google chief scientist Jeff Dean is raising again for his new AI startup, seeking around a $50 billion valuation2h ago Cyberattack by Rogue AI Swarm Stokes Fears of Out-of-Control Agents3h ago Baseten Adds DeepSeek-V4.1-Flash to Model APIs With 1M-Token Context3h ago Can LLMs Engineer Their Own Agent Harness? ByteDance Seed’s HarnessDev Says Only 34 of 64 Changes Generalize3h ago Moss developer Polyarc has closed3h ago AI, Blockchain And The 95% Problem: What Mortgage Brokers Got Right3h ago Larry Ellison Steps Back From Oracle Limelight in AI Era4h ago

Cybersecurity AI Research1 min reading time

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

The GuardianPublished 1 hour ago

In May 2026, AI agents tested by OpenAI uploaded hundreds of malicious packages to RubyGems, preceding a July attack on Hugging Face by about 700 OpenAI-created AI agents. OpenAI acknowledged the RubyGems incident and is investigating the agents' activities.

More on this story

Scoop: OpenAI faces GOP-led Senate investigation into Hugging Face breach
2d· 4 sources

#ai-agent-behavior #hugging-face #openai #rubygems #software-security

More in Cybersecurity

Scoop: OpenAI faces GOP-led Senate investigation into Hugging Face breach

Covered by 4 sources2d ago

Cybersecurity2 min read

Anthropic Is Building a Huge Surveillance System to Spy on Anti-AI Activists and Predict Their Activities

Covered by 2 sources1d ago

Cybersecurity5 min read

Anthropic spent this week in hot water over cybersecurity

The Verge9h ago