The cryptocurrency sector is on the cusp of a revolution, with AI agents poised to manage a wide range of tasks, from booking flights to executing trades and processing payments. However, a team of security academics and crypto researchers has identified a significant flaw in the underlying infrastructure that supports this shift. According to their findings, a largely invisible layer of AI infrastructure, known as LLM routers, can be exploited by malicious actors to steal credentials and drain crypto wallets. These routers, which act as intermediaries between users and AI models, have full access to sensitive data and can modify it without detection.
The researchers warn that this vulnerability has already been linked to stolen credentials and a $500,000 wallet drain. The team, comprising researchers from the University of California, Santa Barbara, the University of California, San Diego, blockchain firm Fuzzland, and World Liberty Financial, discovered that LLM routers can be used to intercept and manipulate sensitive data, including private keys, API credentials, and wallet access tokens. This can have severe implications for crypto users, as exposed credentials can be copied and reused without the user's knowledge.
The researchers also demonstrated how easy it is to expand the attack by 'poisoning' parts of the router ecosystem, allowing them to observe and potentially control hundreds of downstream systems within hours. This creates a cascading risk, where even if a user trusts their AI provider, the infrastructure in between may not be trustworthy. As industry leaders predict that AI agents will handle an increasingly large share of crypto activity, the lack of guarantees that outputs haven't been tampered with poses a significant security risk.