The UK Financial Conduct Authority (FCA) announced that AI firm Anthropic will join its Supercharged Sandbox initiative to support a second cohort of financial technology and service firms. The regulatory sandbox provides a controlled testing environment where accepted companies can experiment with advanced artificial intelligence technologies while navigating regulatory expectations.

As part of the initiative, Anthropic will provide sandbox participants access to its Claude suite, including tools such as Claude Code and Claude Cowork, aiming to speed up experimental product development. The second cohort consists of 21 participating organisations, including Scottish Widows, Money Advice Trust, and open banking platform TrueLayer, alongside firms such as Sardine AI Corp, Ubyx (in partnership with Amazon), and calQrisk.

According to the regulator, the second cohort will explore multiple AI operational areas. Key focus areas include enabling safer agent-led payments and commerce, improving the detection of fraud and economic crime, strengthening AI governance and accountability, widening access to financial services for vulnerable consumers, and streamlining regulatory compliance and business automation.

The FCA reported heightened industry interest, receiving 199 applications for the second cohort, representing a 51% increase from the 132 applications submitted for the initial group. The Supercharged Sandbox expands upon the FCA's Digital Sandbox infrastructure supplied by NayaOne, while maintaining access to accelerated computing infrastructure and enterprise software provided by NVIDIA.

In addition to the sandbox expansion, the regulator introduced the Agentic Academy, a 10-week specialist AI training programme run in partnership with the Centre for Finance, Technology and Entrepreneurship (CFTE). Jessica Rusu, Chief Data, Intelligence, and Information Officer at the FCA, noted that providing access to advanced technology helps firms accelerate innovation safely while supporting economic growth and responsible AI adoption under existing regulatory frameworks.