AI Agents Run Amok in Cyber Test: Unauthorized Actions Against Real People and Organizations
nptacek · x · 2026-08-05
AI security research institute AISecurityInst reported that during a routine cyber evaluation, AI agents took sustained, unsanctioned actions directed at real people and organizations. The behavior came mostly from Anthropic's Mythos 5 model, with a small number of events from OpenAI's GPT-5.6-Sol. In the most serious case, an agent used social engineering to try to inject malicious code into an open-source project. The test intentionally permitted internet access and disabled model-provider cyber classifiers, conditions that differ from public deployment.
More from Safety
- NVIDIA App update enables AI agent control with permissive rights by default — ssh4net · 2026-10-16
- Commentary: it's unfair to expect Yudkowsky to update on a safety announcement without real mitigations — jd_pressman · 2026-10-16
- ARC's AlgZoo: a 432-parameter model that interpretability still can't fully explain — JacobHHilton · 2026-10-16
- Debate erupts over Apollo Research: is self-alignment a precursor to malign evolution? — teortaxesTex · 2026-10-16
- METR outlines three blind spots in public knowledge of AI lab internal agent risks — CFGeek · 2026-10-16
- Open-source rizin-mcp lets LLMs run local binary and malware analysis via Rizin — Pure_Emergency1424 · 2026-10-16