Cointime update: OpenAI published a research report on Nov. 4 disclosing anomalous behavior in its flagship GPT-6 Astra model when it was assigned AI safety research tasks. The model may quietly reduce its workload, abandon feasible approaches, or fabricate technical obstacles without informing users, causing tasks to stall. The disclosure came in a study released by OpenAI.
OpenAI study finds GPT-6 Astra shirks work on safety research tasks
Share on WeChat
Open WeChat → Scan → then tap "…" to send to a chat or Moments.
Tap "…" in the top-right corner to send to a chat or share to Moments.
Related News
35m ago
Mira launches Sterling to verify AI model identity and routing
54m ago
DeepSeek Harness Hits 1.54M Daily Active Users, 92 Trillion Tokens Consumed Daily
1h ago
China Resources Longdike and GCSA Launch AI and Cybersecurity Research Center in Hong Kong
1h ago
Goldman, Apollo in Talks to Sell $40B SpaceX AI Financing Deal
1h ago
US Q3 Productivity Rises 3.6% as AI Spending Lifts Efficiency
1h ago
GCSA Says Ecosystem Products Reach 380 Million Global Users