**UK Banks Face Cybersecurity Risks Amid Restricted Access to Anthropic’s Mythos**
UK banks are confronting significant cybersecurity challenges due to their limited access to Anthropic’s advanced AI model, Mythos. This situation has prompted calls for the UK to enhance its own artificial intelligence capabilities and adopt a more strategic approach to AI technology. Harriet Rees, the Chief Information Officer of Starling Bank and the government-appointed “AI champion,” emphasized the urgency of this need during a recent interview.
Mythos, which was released by the US-based company Anthropic in April, is regarded as one of the most sophisticated AI models for identifying cybersecurity vulnerabilities. It enables financial institutions to quickly patch weaknesses and bolster their defenses. However, access to Mythos has been tightly controlled by US regulations, with only a select group of organizations, primarily US lenders, granted permission to utilize the technology. As a result, major British banks are still awaiting access, with no clear timeline for when they might be able to leverage this critical tool.
Rees highlighted the importance of developing AI infrastructure and skills within the UK to reduce reliance on American technology. “Now is the time for us to think very strategically about what we need to do to protect our leading position moving forward. Time really is of the essence; we don’t have two years here,” she stated.
The restrictions on Mythos access were reinforced by the US administration, which has implemented strict guidelines on who can utilize the AI model. Although Anthropic has begun to roll out Mythos to select organizations outside the United States, the pace of this expansion remains uncertain. An Anthropic spokesperson confirmed that they are working closely with the US government to broaden access to both domestic and international partners.
In a related context, Bank of England Governor Andrew Bailey recently reiterated the necessity for international cooperation in addressing the risks posed by advanced AI models. During his Mansion House speech, he stressed that no country can isolate itself from the global implications of these technologies, underscoring the need for a coordinated approach.
To address the challenges posed by AI in the banking sector, Rees and Rohit Dhawan, head of AI at Lloyds Banking Group, have developed a comprehensive package of recommendations aimed at enhancing AI policy and regulation. Their proposals, which were published as part of the government’s financial services AI adoption plan, advocate for a review of the increasing use of AI chatbots in delivering financial advice to consumers. This initiative comes ahead of finance minister Rachel Reeves’ annual Mansion House speech, where she outlined her vision for AI sovereignty in the UK.
Reeves expressed her commitment to supporting British companies in competing effectively within the global market, stating that the AI Adoption Plan is crucial for achieving this goal in the financial services sector. The UK government plans to collaborate with regulators and industry stakeholders to implement the next steps outlined in the recommendations.
Regulators have long cautioned about the risks associated with the concentration of critical technology providers, which could threaten financial stability and operational resilience. To mitigate these risks, Rees suggested that the UK should not only focus on boosting domestic AI capabilities but also seek partnerships with AI companies outside the United States, including those in China and France.
Furthermore, Dhawan proposed that the UK could enhance transparency and resilience among AI companies that banks currently rely on by designating them as “critical” providers to the finance sector. This designation would allow financial regulators to oversee these companies more closely, similar to the recent announcement regarding four major US cloud providers.
As the landscape of AI technology continues to evolve, the UK banking sector faces pressing challenges in cybersecurity and technology access. The ongoing discussions and recommendations from industry leaders and government officials highlight the need for a proactive and strategic approach to ensure that the UK remains competitive and secure in the face of rapidly advancing AI capabilities.