Teaming up with... AVIVA

Welcome to the UKGI weekly regulation update service for Aviva ABC brokers

We hope you find the Updates useful. If you are
interested in subscribing to our affordable
ABC compliance support package, please
email us at ABC@ukgigroup.com or
call UKGI on our dedicated ABC
contact line 01925 767893.

FCA and Bank of England conduct fourth survey of AI and machine learning in UK Financial services

Date Issued:26/10/2026Applicable to:All firms
Link(s):  Artificial intelligence in UK financial services – 2026 | Bank of England – the UK’s central bank
AI survey places emphasis on supplier oversight

Context

The FCA and Bank of England conducted a fourth survey of artificial intelligence and machine learning in UK Financial Services. The survey highlights the growing use of AI and machine learning in the UK Financial Services market which should encourage firms to review how suppliers deploy AI tools when delivering services to them.

Key points to note and next actions

  • Third-party AI systems reliance is increasing: External providers supply 39% of AI systems (up from 33% in 2024), and they accounted for more than half of the high-materiality AI systems in operation (53%, up from 35% in 2024).
  • Supplier dependence creates risks: Some systems are difficult to replace, which creates concerns around information security, operational resilience, transparency and negotiating contractual protections with providers.
  • AI adoption is widespread: 86% of UK financial services firms use AI across functions including operations, IT, risk, compliance, marketing, HR and legal (up from 75% in 2024). Operations and IT accounted for approximately 31% of all reported AI systems. Most tools used are generative AI.
  • AI agents are growing, but autonomy remains limited: 80% of firms already use agents and another 14% plan to introduce them within a year. They mainly support decision-making and fewer than 10% expect agents to take specified higher-impact actions within the next 12 months.
  • Governance needs to demonstrate control: Boards and senior management need visibility of where AI operates, what it influences, its data and supplier dependencies, accountability and whether outcomes meet expectations. 
  • Oversight must cover future changes: Firms need effective oversight and contractual provisions for advance notification and information about significant model changes that could introduce new risks.