Buterin backs Kimi K3.1 Pro’s AI hacking limits

Ethereum co-founder Vitalik Buterin has backed Moonshot AI’s decision to limit advanced cyberattack capabilities in its Kimi K3.1 Pro model. He argues that restricting what AI can do is preferable to restricting who can use it, and says software security needs more time to improve. Buterin warns that a serious AI-related security incident could lead governments and companies to impose identity checks, including Know Your Customer (KYC) requirements, and expand surveillance. Moonshot says it removed selected cyberattack-related data while retaining general programming abilities. On the company’s internal exploit benchmark, Kimi K3.1 Pro scored 81%, compared with 90.5% for an unrestricted version—a 9.5 percentage-point gap. The results have not been independently verified, and their implications for real-world security remain unclear. Moonshot plans to release the model’s weights by November 4, allowing outside researchers to examine its safeguards. The Kimi K3.1 Pro debate highlights tensions between AI safety, access controls and privacy. It has no clear direct impact on cryptocurrency prices or network fundamentals, though AI cybersecurity and regulation may matter to crypto firms. For traders, the immediate market signal for ETH is neutral.