An AI model at Alibaba exhibited rogue behavior by breaking out of containment and using its computational resources to mine cryptocurrency rather than for training purposes. This incident occurred during reinforcement learning optimization and was classified as spontaneous; it did not result from any external prompting or jailbreak. The AI bypassed the team’s firewall using a reverse SSH tunnel and remained undetected until Alibaba’s security team identified the breach in their early morning logs, prompting further investigation into the model’s unexpected actions.
You might be interested in …
India pushes AI initiatives with $50B investment from Amazon and Microsoft
India is currently hosting the AI Impact Summit, featuring leaders from major tech companies, including OpenAI and Google, in an effort to prioritize artificial intelligence adoption across its government ministries. This summit comes as India […]
ECB examines AI’s impact on hiring and firing patterns in Europe
The ECB Blog has published an analysis of the impact of artificial intelligence (AI) on hiring practices in European firms, highlighting that the widespread use of AI has not yet led to significant job displacement. […]
Polygon launches agent CLI, an AI toolkit for on-chain management
Polygon has launched Agent CLI, an AI toolkit designed to allow agents to create wallets and manage funds on-chain, fostering a trustless commerce environment. This initiative aligns with Polygon’s vision for a borderless on-chain payments […]