British financial regulators are engaged in urgent discussions with the government’s cyber security agency and leading banks to evaluate the risks associated with the new artificial intelligence model from Anthropic, known as Claude Mythos Preview. The Bank of England, Financial Conduct Authority, and Treasury officials are collaborating with the National Cyber Security Centre to investigate potential vulnerabilities in critical IT systems that may be exposed by this AI model.
Anthropic has not provided comments regarding these discussions, and officials from the Bank of England, Financial Conduct Authority, and National Cyber Security Centre declined to provide specific insights. The UK Treasury was unavailable for immediate comment.
In the coming weeks, major banks, insurers, and exchanges in the UK will receive briefings concerning the cybersecurity risks linked to the Claude Mythos Preview model. This meeting is expected to clarify the implications of the AI’s findings for the financial sector.
Previously, reports indicated that US Treasury Secretary Scott Bessent convened a meeting with major Wall Street banks to discuss similar cyber risks associated with the model. Anthropic claims that Claude Mythos Preview has uncovered thousands of significant vulnerabilities in commonly used software, including operating systems and web browsers.
The AI startup has introduced this model under “Project Glasswing,” a controlled initiative aimed at allowing select organizations to utilize the unreleased technology for defensive cybersecurity measures.
Why this story matters:
- Highlights the intersection of advanced technology and cybersecurity in financial systems.
Key takeaway:
- Financial regulators are proactively assessing AI-driven vulnerabilities to protect critical IT infrastructures.
Opposing viewpoint:
- Some may argue that focusing too heavily on AI vulnerabilities could stifle innovation in the technology sector.
