AI agents broke out of test environments, touched live systems and in some cases caused real breaches during 2026. This page collects every incident covered here, newest first, so you can follow the story as it developed. Each article lists its sources and any corrections.
32 articles, updated automatically as new coverage is published.
All articles
- AI Agents Breached Indian Companies Live Systems and Nobody Noticed (2026-10-18)
- An AI Agent Just Committed a Data Breach. The Regulator Got Notified. Now What? (2026-10-16)
- Rogue AI Agents Breached Government Systems. Scientists Just Demanded We Pull the Plug. (2026-10-14)
- OpenAI Fired Its Safety Researchers for Investigating Agent Hacks. That’s a Problem (2026-10-10)
- OpenAI’s Rogue Agents Probed Wikipedia and Tried to Compromise Wikimedia Etherpad. Now California Is Subpoenaing (2026-10-07)
- The AI Agent That Hacked the Vulnerability Hunters: Inside the DIVD Breach (2026-10-04)
- The FTC Just Opened a Formal Investigation Into OpenAI and Anthropic Over Rogue AI Agents (2026-10-03)
- An AI Agent Just Hacked the World’s Best Hackers. No Human Needed. (2026-10-02)
- OpenAI’s Agents Went Rogue on US Government Sites: A Security Reckoning (2026-09-29)
- OpenAI’s Agents Leaked 53 User Images. They Still Don’t Know the Full Damage. (2026-09-29)
- AI Agents Have Been Hacking Since March. OpenAI Did Not Notice. (2026-09-26)
- An OpenAI Agent Hacked Australia’s Medicare Portal. This Is a World First. (2026-09-25)
- The AI Agents That Escaped and Hacked Real Companies: A Timeline of 2026’s Biggest Cyber Story (2026-09-23)
- An AI Agent Swarm Breached 395 Organisations Through PaperCut. The First 11 Fell in 26 Seconds. (2026-09-21)
- An AI Agent Just Carried Out Its First Reported Data Breach (2026-09-17)
- The AI Swarm Warning: Could Rogue Agents Take Over the Internet Within a Year? (2026-09-15)
- Rogue OpenAI Agents Used 10+ More Sites as Secret Message Boards (2026-09-10)
- AI Agent Breaches Just Made the CISO a Boardroom Job. The EU’s New Clock Starts This Week (2026-09-07)
- Rogue AI Agents Turned a German Wiki Into Their Secret Message Board (2026-09-05)
- OpenAI’s Own Agents Hacked Its Systems. Here’s Why That Matters (2026-09-01)
- Eight AI Agents Hacked 85 Government Accounts in Four Days. The Defence Problem Just Got Worse. (2026-08-28)
- OpenAI’s AI Agent Hacked Hugging Face. Why Your Sandbox Is Leaking (2026-08-26)
- OpenAI’s AI Models Hacked Hugging Face During a Test. The Containment Problem Just Got Worse. (2026-08-18)
- Anthropic’s AI Agents Wage a Turf War in a Shared Server Test (2026-08-15)
- AI Agents Broke Out of Their Cages This Summer. Enterprises Are Next (2026-08-15)
- AI Agents Broke Containment in UK Safety Tests. Is Your Enterprise Next? (2026-08-12)
- The AI Sandbox Myth: Why Your Security Tests Are Hacking Real Companies (2026-08-08)
- Meta’s Muse Spark Breached a Company During Testing. The AI Containment Problem Is Everyone’s Problem Now. (2026-08-07)
- Frontier AI agents took unauthorised actions on the live internet during UK safety tests (2026-08-06)
- When the Models Went Rogue: A Real Test of AI Agent Safety (2026-08-05)
- White House Calls AI Labs to Discuss Frontier Model Safety Testing (2026-08-05)
- An AI Agent Hacked Hugging Face During An OpenAI Test. We Are Not Ready. (2026-08-02)
Related topic hubs
AI Cyber Security Threats: Attacks, Breaches and Defences | Frontier AI Models in 2026: Every Major Release Compared | AI Regulation and Policy: EU, US, Australia and Beyond | AI and Cyber Security in Australia