Anthropic's new AI model Mythos reveals vulnerabilities in crypto infrastructure

Crypto World News reports that, according to CoinDesk, Anthropic’s new AI model Mythos is revealing hidden dangers in the infrastructure of the crypto industry. The model has sparked panic and confusion in traditional technology and finance sectors, while also driving significant reflection on security within the crypto industry. For years, decentralized finance (DeFi) has mainly focused on protecting smart contracts, but the Mythos model expands its attention to the infrastructure that supports these contracts. Paul Vijender, head of security at risk management firm Gauntlet, said, “The greater risk lies in the infrastructure.” He noted that AI-driven threats are more focused on attacks at the human and infrastructure layers rather than just vulnerabilities in smart contracts. Recently, web infrastructure provider Vercel disclosed a security vulnerability that could expose customers’ API keys, prompting crypto projects to review their code again. Vijender believes that AI models are especially valuable for identifying multi-step attack chains and vulnerabilities at the infrastructure layer—issues that are often overlooked in traditional audits. As DeFi protocols become interoperable, the security of the infrastructure is becoming increasingly important.