Barclays has been ushered into the second cohort of firms handpicked by the City watchdog for its artificial intelligence live testing scheme, as Britain’s banking establishment doubles down on the technology race reshaping financial services.
The FTSE 100 lender will rub shoulders with its high street rival Lloyds Banking Group, which is entering the programme through its Scottish Widows subsidiary, alongside credit reference agency Experian, payments outfit GoCardless and Swiss banking giant UBS.
Run by the Financial Conduct Authority in partnership with Advai, the British specialist in automated testing, evaluation and assurance of AI systems, the initiative offers successful applicants a regulatory safe harbour in which to put their models through their paces. The intention is to let firms iron out governance wrinkles well before those systems are turned loose on high-stakes decisions affecting consumers.
Speaking at Innovate Finance’s Global Fintech Summit, Jessica Rusu, the FCA’s chief data, information and intelligence officer, said the scheme “reflects our commitment to supporting the pace of change in AI, whilst demonstrating how regulators and industry can work together to harness innovation responsibly”.
The announcement lands at a moment when Britain’s traditional lenders are under acute pressure to demonstrate tech credentials that can stand comparison with the tech-native neobanks snapping at their heels. Investors have grown increasingly impatient for a convincing AI narrative, particularly one that sets out concrete implications for costs and headcount.
UBS analysts warned earlier this year that banks would be “pressed hard” to articulate a “coherent financial story for AI implementation: what is being spent now and what it means for the future shape of expenses overall and headcount in particular”.
The urgency is reflected in the flurry of alliances struck in recent months. Barclays has thrown in its lot with Microsoft AI in a deal that will put AI tools in the hands of some 100,000 of its bankers, while NatWest has signed with OpenAI and HSBC has turned to the French champion Mistral. NatWest, Lloyds and HSBC each sit within the top 20 of the Evident AI index, the global benchmark for AI adoption in banking.
Yet for all the enthusiasm, the risks have not gone unnoticed. American regulators recently summoned Wall Street chief executives to an emergency meeting amid concerns that Anthropic’s newly released “Mythos” tool could pose systemic risks to the financial system, a reminder that the cybersecurity implications of ever more capable models remain a live worry for supervisors on both sides of the Atlantic.
The FCA launched its first AI live testing cohort last December, with Monzo, NatWest and Santander among the inaugural participants. For smaller and mid-market firms watching from the sidelines, the expanding programme offers a useful weathervane on where the regulator will draw its lines as AI embeds itself deeper into British finance.
Read more:
Barclays and Lloyds join FCA’s second AI sandbox as banks race to prove their tech credentials