Anthropic backs FCA AI sandbox

Anthropic will support the second cohort of firms taking part in the FCA's Supercharged Sandbox.

The scheme gives firms access to Claude, Claude Code and Claude Cowork to help speed up their AI development work.

The second cohort includes 21 firms, among them Scottish Widows, Money Advice Trust and TrueLayer, selected after a 51% rise in applications compared with the first group.

The Supercharged Sandbox gives firms a controlled environment to safely experiment with advanced AI, building on the FCA's existing Digital Sandbox run with NayaOne.

Jessica Rusu, chief data, intelligence and information officer at the FCA, said: "The high level of interest in the Supercharged Sandbox demonstrates the demand for trusted environments where firms can experiment safely and responsibly."

The 21 firms will explore a range of AI use cases, including solutions designed to enable safer agent-led payments and commerce, detect fraud and economic crime more effectively, strengthen AI governance and accountability, widen access to financial services for vulnerable and underserved consumers, and streamline compliance and business automation.

The Sandbox builds on support already in place from NayaOne and NVIDIA. NVIDIA provides accelerated computing infrastructure and AI Enterprise software to participating firms.

The FCA has also launched a separate initiative: the Agentic Academy, a 10-week specialist AI programme for selected firms delivered jointly by the FCA and Centre for Finance, Technology and Entrepreneurship.