This week’s dominant theme is accountability and governance: the FCA’s landmark Mills Review published on 6 July sets the tone for how AI will be regulated in UK retail finance through 2030, while the EU AI Act’s August high-risk deadline intensifies compliance pressure on banks and insurers across Europe. Alongside the regulatory push, a disruptive new insurtech startup is threatening traditional broker distribution by embedding insurance sales directly inside AI chatbots — signalling that the AI-native distribution war in financial services has begun in earnest.
Top story: The FCA published the Mills Review on 6 July — the first regulator-initiated AI study of its kind globally — setting a 2030 roadmap for AI in retail financial services and confirming no new AI-specific rules, but a materially higher governance bar.
FCA’s Landmark Mills Review Maps AI in Finance to 2030
FCA (Official) · Regulation
Published on 6 July 2026, the Mills Review is the first regulator-initiated AI study of its kind globally, drawing on 140 submissions and a survey of 5,026 UK adults. It identifies four major AI-driven shifts reshaping retail financial services — from firm operations to fraud amplification — and makes seven recommendations, notably a 3–6 month perimeter review of how consumers use general-purpose LLMs like ChatGPT for financial decisions, and a proposal for a free public AI-powered financial capability service. Critically, it confirms no new AI-specific rulebook; instead, the Senior Managers Regime and Consumer Duty remain the anchors — but with a significantly higher bar on evidence, oversight, and governance.
https://www.fca.org.uk/publications/corporate-documents/mills-review
EU AI Act High-Risk Deadline Hits Banks in 25 Days
Bratby Law · Regulation
With the EU AI Act’s 2 August 2026 enforcement deadline for high-risk AI systems now imminent, banks and insurers using AI for credit scoring, AML, loan approvals, and insurance underwriting must meet strict obligations including risk management systems, technical documentation, and human oversight — or face fines up to €35 million or 7% of global turnover. The European Banking Authority has confirmed that the majority of AI use cases at supervised institutions fall into the high-risk category, making this a mainstream compliance challenge, not an edge case. UK-based fintechs serving EU customers are also in scope, regardless of where they are headquartered.
Waniwani Raises $8M to Sell Insurance Inside AI Chatbots
Tech Funding News · Finance
Paris and San Francisco-based startup Waniwani has raised $8M in seed funding led by Seedcamp to let financial services companies sell products directly inside AI platforms like ChatGPT. The startup — whose launch reportedly sent insurance broker stocks down 9% in a single day — was co-founded by the creator of Luko, France’s largest online home insurer before its Allianz acquisition. The raise signals that the AI-native distribution model is becoming a serious structural threat to traditional broker and comparison-site channels in insurance and financial services.
https://techfundingnews.com/waniwani-raises-8m-seed-ai-insurance-chatgpt-app-insurance-brokers/
One-Third of UK Consumers Now Using AI for Pension Decisions
FF News · Strategy
Consumer research tied to the FCA’s Mills Review found that one in five UK adults — equivalent to 11 million people — are open to AI making financial decisions autonomously on their behalf, with demand strongest for pensions, debt advice, and investments. Around 26% already trust general-purpose tools such as ChatGPT or Claude for financial advice, frequently without understanding they have no formal route to redress if things go wrong. Industry experts warn that AI is “excellent at sounding authoritative, but not always at being right” — raising acute product liability and consumer duty questions for pension providers and wealth managers.
BIS Warns AI Shadow Borrowing Is Exposing Insurers and Private Credit
Insurance Journal · Risk
Bank for International Settlements officials have warned that AI hyperscalers are increasingly financing infrastructure expansions through off-balance-sheet arrangements — described as “shadow borrowing” — in partnership with private credit firms and insurers. These structures, typically using special purpose vehicles backed by chip or real estate assets, are creating opaque links between Big Tech and non-bank investors, with banks also providing funding lines that could become new shock transmission channels. With global data centre spending projected to reach $7 trillion by 2030, the BIS cautions that the scale and lack of transparency in these financing structures echoes historic credit bubble conditions.
https://www.insurancejournal.com/news/international/2026/03/17/862128.htm