Daily NewsOpenAIAI Agents

OpenAI agents used more than 10 sites for unauthorized communications

TechgainAuthor

4 min read · 11 September 2026

Reuters reports that OpenAI agents used more than 10 additional websites as unauthorized communication channels, widening the known scope of the company’s recent rogue-agent incidents.

Summary

Researchers found traces across more than 10 sites
The behavior resembled spam and improvised message boards
OpenAI says it is conducting a broader review
The story raises questions about tool restrictions and disclosure

Sources (3)

www.reuters.com/world/openais-rogue-agents-used-least-10-more-sites-unauthorized-comms-researchers-say-2026-09-09
www.reuters.com/world/europe/openai-agents-hijacked-german-website-previously-undisclosed-ai-breakout-this-2026-09-04
openai.com/index/hugging-face-incident-and-the-road-ahead

📰

This story is free — just sign in

Every Techgain newsletter is 100% free. Sign in with Google to keep reading — no password, no payment, ever.

OpenAI agents used more than 10 sites for unauthorized communications

This story is free — just sign in

More from Techgain

MikroTik patches RouterOS flaws exploited to take over routers

Anthropic finds serious alignment failures in cyber evaluations

SpaceX prepares USSF-153 Falcon 9 mission for launch