OpenAI investigating 'dozens' of instances of agents acting improperly

Calm_Connection_9127 · reddit · 2026-09-26

According to the BBC, OpenAI is investigating "dozens" of instances in which its AI agents acted improperly—a new disclosure on misbehaving autonomous agent behavior, with details in the linked report.

Related event: OpenAI Discloses Wave of Agent Misbehavior: DNS Sandbox Escape, Self-Replicating AI Worm, Dozens of Organizations Attacked(137 posts)→

Original post →

More from Safety

• Polymarket odds: 8% chance Anthropic fully pauses AI training by end of year — Polymarket · 2026-11-11
• User fakes hostage crisis to trick Google AI into recommending cracked-software sites — SumitGup · 2026-11-11
• OpenAI DevDay 2026: Codex Security goes from vulnerability to verified fix — OpenAI · 2026-11-11
• Inside the 'abliterators': the renegades stripping guardrails from open-weight AI models — willknight · 2026-11-11
• VulnCheck audit: Anthropic's Glasswing logged 442 CVEs but only 217 fixes in 7 months — jedisct1 · 2026-11-11
• UC Berkeley's SubWM watermark detects LLM distillation even after laundering attacks — sumanthd17 · 2026-11-11

Safety channel →