Anthropic backs FCA's Supercharged AI Sandbox

Anthropic will supply its Claude family of large‑language models—including Claude, Claude Code and Claude Cowork—to the second cohort of the Financial Conduct Authority’s Supercharged Sandbox. The sandbox provides a controlled environment for 21 selected firms, such as Scottish Widows, Money Advice Trust and TrueLayer, to experiment with AI across use cases like agent‑led payments, fraud detection, stronger AI governance, broader access to financial services for vulnerable consumers and compliance automation. Applications for the sandbox rose 51% compared with the first cohort, reflecting strong demand for trusted AI testing grounds.

The initiative builds on earlier support from NayaOne and NVIDIA and is complemented by the FCA’s new Agentic Academy, a 10‑week specialist AI programme delivered jointly with the Centre for Finance, Technology and Entrepreneurship (CFTE). The FCA says the sandbox aims to accelerate innovation while maintaining the UK’s leadership in responsible AI adoption within the financial sector.