• Limited UK access to Anthropic's Mythos model has sparked calls for greater domestic AI capability
• Government AI Champion Harriet Rees warned Britain must reduce reliance on U.S. technology providers
• Banks increasingly view frontier AI as essential for cybersecurity and operational resilience
• The Bank of England has called for international cooperation on supervising advanced AI models
• Industry proposals include investment in AI infrastructure, skills and regulatory collaboration



Britain's has been warned it must resist the temptation to allow unchecked dependence on overseas AI tech amid fears it could damage the long-term health of the UK banking sector.

The country’s inability to secure widespread access to one of the world's most advanced artificial intelligence models has intensified calls for a national strategy to strengthen domestic AI capabilities.

Industry leaders have warned that overreliance on foreign technology could weaken the long-term competitiveness and resilience..

The debate has been triggered by the limited availability of Anthropic's frontier AI model, Mythos, which has been made available to only a small number of financial institutions, primarily U.S. banks and selected organizations.

The model has attracted significant attention because of its ability to identify software vulnerabilities rapidly, helping institutions strengthen cyber defenses while also raising concerns that malicious actors could exploit similar capabilities.

Harriet Rees, Chief Information Officer at Starling Bank and the Treasury's appointed AI Champion for financial services, said the situation should serve as a wake-up call for Britain.

"Now is the time for us to think very strategically about what we need to do to protect our leading position moving forward," she said. "Time really is of the essence, we don't have two years here."

Rees argued that Britain must invest in its own AI infrastructure, foundation models and specialist skills to reduce dependence on U.S. technology providers.

She warned that relying entirely on overseas companies for strategically important AI capabilities could place the UK's financial services industry at a competitive disadvantage as banks increasingly integrate advanced AI into cybersecurity, operational resilience and fraud detection.

The concerns extend beyond commercial competitiveness. Risk leaders increasingly view frontier AI as both a defensive tool and a potential threat.

Models capable of discovering software weaknesses can help banks identify vulnerabilities before attackers exploit them, but they may also significantly reduce the time available to detect and respond to sophisticated cyber threats.

That risk has already prompted regulatory attention elsewhere.

Canada's Office of the Superintendent of Financial Institutions recently warned banks and insurers that advanced AI models could accelerate cyberattacks against legacy technology environments, urging firms to strengthen governance and response capabilities.

The Bank of England has also highlighted the growing strategic importance of frontier AI. Governor Andrew Bailey has called for greater international cooperation on testing and supervising advanced AI models, arguing that no single jurisdiction can effectively manage the associated risks in isolation.

Speaking ahead of the Mansion House dinner, Bailey said collaboration between governments, regulators and technology companies would be essential as AI capabilities continue to advance, particularly where cyber resilience and financial stability are concerned.

Rees' comments accompany a broader package of recommendations developed with banking executives and submitted to the UK government as part of its AI Adoption Plan.

The proposals call for increased investment in domestic computing infrastructure, stronger collaboration between regulators and industry, improved access to frontier AI technologies and greater support for developing AI skills across financial services.

Industry observers say the debate also illustrates the increasing convergence of technology policy and financial regulation.

As AI becomes embedded within critical financial infrastructure, governments are placing greater emphasis on technological sovereignty alongside traditional concerns such as capital, liquidity and cybersecurity.

For Britain's banking sector, the challenge is no longer simply adopting artificial intelligence quickly enough.

It is ensuring that access to strategically important technologies remains sufficiently diverse, resilient and domestically supported to avoid creating new forms of dependency at a time when digital resilience has become a cornerstone of financial stability.