The rapid adoption of AI agents in the cryptocurrency industry, predicted to handle trillions of dollars in transactions by 2030, may be hindered by a significant security flaw. According to a recent study, a largely overlooked component of AI infrastructure can be exploited by malicious actors to intercept sensitive information, including private keys and wallet access tokens. Researchers have identified a weakness in LLM routers, which act as intermediaries between users and AI models, allowing them to access and modify sensitive data. This vulnerability has already been linked to stolen credentials and a $500,000 crypto wallet drain.
The researchers found that these routers can be used to steal credentials, drain wallets, and even take control of hundreds of downstream systems. The implications are severe, with private keys, API credentials, and wallet access tokens often passing through these systems in plain text.
The team 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 systems within hours. The study highlights a weakest-link problem, where a single malicious router in the chain is enough to compromise the entire system, creating a cascading risk for users who trust their AI providers but may not be able to trust the infrastructure in between.