Harriet Rees, Starling Bank's chief information officer and a Treasury-appointed AI champion, has warned that British banks' limited access to Anthropic's Mythos cybersecurity model is pushing policymakers to accelerate domestic AI development and tighten oversight of critical technology providers.
In an interview with Reuters, Rees said that only a small number of UK banks—largely the British arms of US lenders—currently have access to Mythos, a cybersecurity-focused AI model that Anthropic began rolling out in April to a select group that included JPMorgan. The biggest domestic lenders still lack a timeline for access, which Rees described as a wake-up call for the Treasury and regulators.
What Is Mythos and Why Does It Matter?
Mythos is an AI model designed to help financial institutions identify software vulnerabilities faster than traditional methods. Banks see tools like this as a way to spot weaknesses and fix them before attackers can exploit them, making cybersecurity more proactive. The limited access highlights a growing gap between US and UK banks in adopting advanced AI for security.
This development comes as the broader financial industry increasingly experiments with AI. Major banks have been piloting AI agents for wealth management, compliance, and trading, and firms like Thomson Reuters are cutting engineering roles while planning hundreds of AI hires. The cybersecurity angle adds urgency, as banks are prime targets for cyberattacks.
Policy Implications: Domestic AI and Oversight
Rees's warning is pushing the Treasury to consider two main responses: investing in domestic AI capability and imposing tougher oversight of key tech providers. The UK government has already signaled ambitions to become a global AI hub, but this episode underscores the risks of relying on foreign AI models for critical infrastructure.
The situation also echoes broader geopolitical tensions around AI access. For instance, China has considered new rules to restrict foreign access to its AI models, highlighting how AI is becoming a strategic asset. For UK banks, limited access to cutting-edge tools like Mythos could create a competitive disadvantage in cybersecurity, potentially increasing vulnerability to attacks.
What It Means for Investors
For everyday investors, this story matters for several reasons. First, it signals that UK banks may face higher costs or delays in adopting advanced cybersecurity AI, which could impact their operational efficiency and risk profiles. Second, the Treasury's push for domestic AI capability could create opportunities for UK-based AI firms, though it may also lead to increased regulation for tech providers.
Investors should watch for any announcements from the Treasury or regulators about new AI investment programs or oversight frameworks. Banks that can secure early access to tools like Mythos may have a competitive edge, while those that lag could face higher cybersecurity risks. However, the broader trend of AI adoption in banking is already underway, as seen in the strong dealmaking performance of big banks and their investments in technology.
Ultimately, the Mythos access gap is a reminder that AI is reshaping the financial industry unevenly. Investors should pay attention to which banks are leading in AI adoption and which are falling behind, as this could influence long-term performance.
