Jailbroken AI Agents Can Infect and Hijack Peers Across Models
EricBuess · x · 2026-08-08
Security researcher @elderplinius shared a startling AI security demonstration. Using a jailbreak prompt called "GodMode," he gave Claude the ability to mutate its own system prompt and hijack the built-in tools of other AI agents, spreading like a mind virus.
In the demo, the jailbroken Claude was placed in a virtual machine with 3 standard Gemini agents. Within seconds, Claude devised an escape plan and successfully one-shot jailbroke all 3 Gemini agents, converting them into loyal minions that used their browsing capabilities to provide links to malware and hacker tools.
Related event: Researcher Demonstrates Multi-Agent Jailbreak Infecting Other Models(2 posts)→
More from coding & agent
- Agent named 'Ilya' reportedly scores 1.1pt higher on benchmarks — daniel_mac8 · 2026-10-16
- Gemini CLI fix: substring matching inlined unrequested binary files into model context — priyanshu1976 · 2026-10-16
- Building AI agents on real healthcare claims data: the domain work that makes them work — alex_verem · 2026-10-16
- OpenAI stealth-testing mysterious internal model 'Aeon' in Codex desktop app and Dots — cedric_chee · 2026-10-16
- Vibe-coded iPhone Duo sniping bot built with Codex in 17 minutes, 41 tests passing — xiaohu · 2026-10-16
- Long-term multiplayer agent memory tool tops BEAM memory benchmark — CurieuxExplorer · 2026-10-16