AI in finance: adoption is high, autonomy is not
A source-backed guide to AI in financial services, covering adoption, third-party dependence, accountability, automation, agentic finance and the difference between widespread use and fully autonomous decisions.
Last reviewed 2026-07-25
How widely is AI used in UK financial services?
The Bank of England and Financial Conduct Authority reported that 75% of responding firms were already using AI in their 2024 survey, with another 10% planning to use it within three years. Adoption is therefore substantial, but it covers a wide range of use cases from analytics and fraud detection to customer service. “Uses AI” does not mean the firm runs autonomous financial decisions.
How autonomous are financial AI systems?
The same Bank and FCA survey found that 55% of AI use cases had some automated decision-making, while only 2% were fully autonomous. That distinction matters because public discussion often collapses assistance, partial automation and autonomy into one bucket. The governance problem changes sharply when a system moves from advising a human to executing a consequential decision without routine human approval.
Why do third-party AI providers matter in finance?
Financial firms increasingly depend on external technology. The 2024 Bank and FCA survey reported that one third of AI use cases were third-party implementations, with concentration among major cloud, model and data providers. That creates a different risk shape: firms may understand their business decision well while having less visibility into an underlying model, data dependency or provider change.
Are consumers ready for agentic finance?
Interest exists, but appetite is not the same as safe deployment. In July 2026 the FCA said research for its Mills Review indicated about a fifth of people were likely to use AI that can act autonomously within pre-set goals for personal finance. That creates opportunities for easier comparison and administration, alongside sharper questions about permissions, fraud, suitability, redress and who is responsible when an agent acts badly.
What does the current evidence say?
- The 2024 Bank of England/FCA survey found 75% of responding financial firms already used AI, 33% of AI use cases were third-party implementations, and only 2% of use cases were fully autonomous. Source
- The FCA reported in July 2026 that about one fifth of people were likely to use agentic AI for personal finance within pre-set goals. Source
Limitations
The Bank/FCA adoption figures describe survey respondents, not every UK financial firm, and firms self-classified aspects such as materiality. Consumer stated likelihood of using agentic finance does not predict actual uptake or outcomes.
A counterpoint worth keeping
High adoption does not imply finance has solved AI governance. The more useful evidence may be the modest level of full autonomy and the continuing importance of accountable people, controls and third-party oversight.
Sources and provenance
- Bank of England and Financial Conduct Authority · Artificial intelligence in UK financial services - 2024 · 2024-11-21 · Primary source
- Financial Conduct Authority · FCA publishes landmark review into impact of AI on retail financial services · 2026-07-06 · Primary source