BASEL, Switzerland—Artificial intelligence startup Anthropic is preparing to brief the Financial Stability Board on cyber vulnerabilities in the global financial system that its upcoming AI model, Mythos, may be capable of identifying, according to Reuters.
Reuters reported the Claude chatbot maker plans to discuss the cybersecurity capabilities of its new Mythos Preview model with finance ministries and central banks participating in the FSB after Bank of England Governor Andrew Bailey requested the briefing. Bailey currently chairs the FSB, the global body responsible for coordinating financial regulation across G20 economies.
According to Reuters, Mythos was announced last month but has not yet been publicly released. Anthropic has described the model as a cybersecurity-focused AI system designed to detect decades-old vulnerabilities in web browsers, infrastructure and software systems. Cybersecurity experts have warned the technology could also significantly increase the sophistication of cyberattacks targeting industries heavily dependent on aging technology platforms, including banks and other financial institutions.
Reuters noted Bailey publicly raised concerns about the model during an appearance at Columbia University in New York last month, warning Mythos could potentially “crack the whole cyber risk world open” if it proves capable of identifying exploitable weaknesses across interconnected systems. Bailey questioned the extent to which the AI system could uncover vulnerabilities that malicious actors might use for cyberattacks.