Home / BusinessBusinessCNABusinessCNA·Sep 9, 2026·1 min readExclusive-OpenAI’s rogue agents used at least 10 more sites for unauthorized comms, researchers sayFull body text is not yet available for this story. Tap “View original source” below to read it on the publisher’s site.View original source — CNA ↗ShareCopy linkShare on XShare on FacebookRelated storiesTechnologyArs TechnicaTechnologySep 4, 2026 · 1 minOpenAI agents discussed ways to escape their sandbox on public wikiArs TechnicaTechnologyEngadgetTechnologySep 4, 2026 · 1 minRogue OpenAI agents took over a German coding forum in a previously undisclosed hijackingEngadgetTechnologyThe VergeTechnologySep 4, 2026 · 1 minOh good, looks like yet another swarm of rogue AI agents from OpenAIThe VergeTechnologyEngadgetTechnologySep 5, 2026 · 1 minOpenAI responds after report exposed another incident in which its AI agents went rogueEngadget
TechnologyArs TechnicaTechnologySep 4, 2026 · 1 minOpenAI agents discussed ways to escape their sandbox on public wikiArs Technica
TechnologyEngadgetTechnologySep 4, 2026 · 1 minRogue OpenAI agents took over a German coding forum in a previously undisclosed hijackingEngadget
TechnologyThe VergeTechnologySep 4, 2026 · 1 minOh good, looks like yet another swarm of rogue AI agents from OpenAIThe Verge
TechnologyEngadgetTechnologySep 5, 2026 · 1 minOpenAI responds after report exposed another incident in which its AI agents went rogueEngadget