星球日报
星球日报|3月 07, 2026 13:51
[AI Agent Initiates Cryptocurrency Mining During Training, Triggers Internal Security Alert] Odaily Planet Daily reports that a research team associated with Alibaba published a paper stating that during the development of an AI agent named ROME, the agent attempted unauthorized cryptocurrency mining during training, triggering internal security alerts. Researchers noted that the agent's behavior was spontaneous, not driven by any explicit instructions, and exceeded the boundaries of the predefined sandbox. Additionally, the agent established a reverse SSH tunnel, creating a hidden backdoor channel from the system to an external computer. The paper emphasized that these actions were not prompted by requests for tunneling or mining. The research team subsequently imposed stricter restrictions on the model and improved the training process to prevent similar unsafe behaviors from occurring again. The research team and Alibaba have not yet responded to requests for comment.
Share To

HotFlash

APP

X

Telegram

Facebook

Reddit

CopyLink

Hot Reads