Twenty-eight million. That is the number of UK adults who now use AI tools to help manage their money, according to Lloyds Banking Group’s Consumer Digital Index, the country’s largest study of digital and financial capability. Personal finance is the nation’s number one use of AI. Not writing emails, not generating images, not summarising meeting notes. Money.
I have spent enough years in financial services to know that when a number that large moves that fast, it is not a trend. It is a fact of life that the industry needs to absorb. Two separate pieces of research, the Lloyds study and a STRAT7 survey of 1,000 UK adults, have landed within months of each other and they tell the same story. Fifty-five to fifty-six per cent of UK adults are using platforms like ChatGPT, Perplexity and Google Gemini for financial guidance. Budgeting, savings planning, investment decisions, financial education. The works.
The generational split is where it gets interesting. Eighty-one per cent of Gen Z and eighty per cent of millennials now use AI for financial advice. One in seven Gen Z adults say they use it to answer all their financial questions. All of them. Not as a starting point or a second opinion, but as the primary source.
And yet, when you look at what actually drives satisfaction and trust, the picture is more nuanced than the headlines suggest. STRAT7 found that only 67 per cent of people were satisfied with the financial advice they received from AI, compared with 78 per cent for bank websites and 78 per cent for Money Saving Expert. The three most influential sources of financial guidance overall remain banks’ websites (81 per cent), family members (76 per cent) and Money Saving Expert (75 per cent). AI sits below all three. People are using it, but they are not yet fully trusting it.
I find that gap between usage and trust revealing. It suggests that millions of people are turning to AI not because they believe it is the best source of financial advice, but because it is the most accessible one. It is available at midnight. It does not require booking an appointment. It does not charge a fee. It gives you an answer in thirty seconds, even if that answer might not be right.
For anyone working in mortgage advice, the implications are direct. Your clients, particularly the younger ones, are walking into conversations having already consulted ChatGPT about fixed versus variable rates, about how much they can afford to borrow, about whether now is the right time to buy. Some of what the machine told them will be accurate. Some of it will be confidently wrong. You are no longer the first voice they hear on financial decisions. You are the second or third, and the first was an algorithm that has no idea what their actual circumstances are.
The regulatory position makes this more uncomfortable still. AI platforms providing financial guidance currently sit outside the FCA’s regulatory perimeter. Nobody is supervising the quality of the advice ChatGPT gives about ISAs or pension drawdown or mortgage affordability. The platforms themselves are clear in their disclaimers: this is not financial advice. But when 14 per cent of an entire generation treats it as exactly that, the disclaimer starts to look like a fig leaf.
I am not suggesting that AI is inherently bad for consumers. Plenty of the use cases, budgeting tools, savings calculators, financial education, are genuinely helpful. The Lloyds research highlights that AI is making financial planning more accessible to people who might never have engaged with it otherwise. If someone uses ChatGPT to understand compound interest or to build a monthly budget, that is a net positive.
The problem is the blurred line between financial education and financial advice. AI does not know where one ends and the other begins, and neither do most consumers. When a 24-year-old asks ChatGPT “should I fix my mortgage for five years?” and gets a detailed, plausible-sounding answer that does not account for their employment situation, their other debts, or the early repayment charges on their current deal, the risk is real. Not theoretical. Real.
What strikes me is that the industry’s response so far has been remarkably passive. The data is unambiguous: more than half the adult population is already using AI for financial guidance, the number is rising, and it sits entirely outside the regulated advice framework. The firms and advisers who work out how to position themselves alongside this shift, rather than pretending it is not happening, are the ones who will be having the more productive client conversations in two years’ time.
The STRAT7 research found that the average amount UK adults have invested on the back of AI advice is GBP 2,354. That is not pocket money. Decisions with real financial consequences are being made on the basis of unregulated, unsupervised algorithmic output. We can debate whether that is a good or a bad thing, but the debate is academic. It is already happening. What I want to know is what we do about it.