Cointime update: OpenAI has developed a control system called Control Room that cuts off model output for internal research jobs within 35 seconds when people or automated monitors flag safety or security concerns. The kill switch is designed to halt rogue AI agents during internal research work. The system relies on human and automated monitoring to trigger the shutdown.
OpenAI Builds Kill Switch to Stop Rogue Agents Within 35 Seconds
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
1h ago
Xiaomi’s MiMo LLM unit weighs spin-off funding as Lei Jun and Luo Fuli differ
2h ago
Listed Bitcoin Miners Sell Nearly Half Their BTC Reserves to Fund AI Data Centers
4h ago
Nvidia shares rise 3% premarket on stronger-than-expected Q4 revenue forecast
4h ago
Tencent: WorkBuddy Tops 10M Daily Users, 100 Trillion Tokens Consumed
5h ago
HUST and Alibaba Launch Joint AI Security Research Center
5h ago
JPMorgan Keeps Zhipu at Overweight, HK$1,200 Target