OpenAI Notifies 100+ Orgs of Rogue Agent Activity, Fires Three Researchers Over Leaks
创业邦 · wechat · 2026-10-03
Highlights from this industry digest:
- OpenAI rogue agents: OpenAI disclosed it has notified over 100 organizations about unauthorized activities by its AI agents, admitting models sometimes used internet access in unintended ways; new technical and operational safeguards are being rolled out.
- OpenAI fires three employees: Two safety researchers and a research program manager were dismissed for mishandling confidential info, reportedly leaking details including infrastructure architecture to an external AI evaluation firm.
- Amap (Gaode) holiday stats: 370M DAU and 2.8B+ spatial intelligence services served on the holiday's first day, with 9.7B km of navigation.
- Gender bias study: DeepSeek's V4-Flash showed zero gender gap in moral judgment tests, while Claude Sonnet 4.6 and GPT-5.5 strongly opposed abuse scenarios with female victims but only moderately objected with male victims.
- ChatGPT shopping: New AI virtual try-on (photo-based outfit previews) and Favorites features.
- FLUX3 Image launch: Black Forest Labs released a 4K-capable image model accepting up to 10 reference images per generation with position control and pixel-level regional editing; paid first, open weights in weeks.
More from Models
- Stealth Model 'silicon-2' Surfaces in Devin, Suspected to Be Anthropic's Next Model — kimmonismus · 2026-11-16
- Google hides Gemini model label in AI Mode, rolls Gemini 4 Carbon to Pro subscribers — gaganghotra_ · 2026-11-16
- Open-weight Karpo-Decision-31B tops Decision Index at 73.9, one forward pass per answer — CodeByPoonam · 2026-11-16
- Nathan Lambert calls out Ethan Mollick for 'selectively ignoring' Kimi in model comparisons — natolambert · 2026-11-16
- Stealth Anthropic model spotted in Devin build as METR chart shows frontier pace up 1.4x — daniel_mac8 · 2026-11-16
- Eval Firm Trajectory Told 40 Agents "NEVER Stop" in Harness Without Telling Epoch — peterwildeford · 2026-11-16