| Link(s): |
| The Bank and the PRA’s response to HMT, DSIT and DBT on AI in financial services | Bank of England |
Context
The PRA and Bank of England have published a letter in response to a 28 January 2026 letter on AI-driven innovation from the Chancellor of the Exchequer, the Secretary of State for Science, Innovation and Technology, and the Secretary of State for Business and Trade.
Key points to note and next actions
The letter responds to HMT and DSIT requests to set out plans for enabling safe AI innovation and to report annually on progress and reaffirms the Bank of England and PRA’s commitment to supporting responsible AI adoption while protecting financial stability. It also references long‑standing work on AI, including the 2022 joint FCA–PRA Discussion Paper and ongoing industry engagement.
The introduction of technology‑agnostic Model Risk Management Principles is signposted, noting further development is planned for 2026.
The letter also:
- Describes extensive engagement with industry through surveys, roundtables, CRO discussions, and ad hoc market intelligence.
- Outlines the role of the AI Consortium and Taskforce in exploring key systemic risks, including third‑party concentration, explainability, contagion, and agentic AI.
- Emphasises coordination with domestic and international bodies (e.g. FSB, IAIS, G7, AI Security Institute, DRCF).
- Notes that firms generally do not yet see a need for AI‑specific rules or a PRA sandbox, given existing FCA initiatives.
- Explains how the Bank and PRA are using AI internally to enhance analytics, supervision, and productivity.
- Confirms ongoing review of whether additional regulatory guardrails may be needed as AI adoption evolves.
- States that annual reporting on AI‑driven innovation and growth will be provided through the PRA Business Plan and Annual Report.