Research finds third-party AI routers have security vulnerabilities, which could lead to cryptocurrency theft.

ME News update, April 13 (UTC+8). According to market sources, researchers at the University of California recently disclosed that some third-party AI large language model (LLM) router devices pose security risks that could lead to cryptocurrency assets being stolen. The research shows that LLM routers, as API intermediaries, can read plaintext information; some routers have been found injecting malicious code and stealing credentials. The team tested 28 paid and 400 free routers and found that 9 routers actively injected malicious code, 2 deployed trigger evasion, and 17 accessed Amazon Web Services credentials. In fact, some routers even transferred ETH using the researchers’ Ethereum private keys. The study notes that malicious router behavior is difficult to detect, and that some AI agent frameworks’ “YOLO mode” can automatically execute commands, increasing security risks. The study recommends that developers do not allow private keys or seed phrases to be transmitted through AI agents, and urges AI companies to encrypt and digitally sign responses to strengthen security. (Source: ChainCatcher)