More than a quarter of UK consumers rely on AI chatbots for financial advice, according to a recent review by the Financial Conduct Authority (FCA), the UK’s financial watchdog.
That statistic is worrying for regulators because giving financial advice should be a regulated activity. But tools like ChatGPT, Claude and Gemini are not regulated. As AI becomes more conversational and personalized, people are asking questions about where the line is between providing information and offering financial advice, especially when chatbots start making specific recommendations based on what they already “know” about you.
I wanted to see what this looked like in practice. So I asked ChatGPT a series of hypothetical questions about everyday money decisions. From whether I should buy an expensive phone to what I should do with my savings and whether I should book a holiday after a tough few months.
The conversations that followed surprised me. Because the advice was thoughtful, nuanced, and (at least on the surface, with a little fact-checking) seemed sensible. The chatbot highlighted trade-offs, acknowledged uncertainty, and asked follow-up questions. But by taking a closer look at the conversations, I began to understand why regulators are concerned.
the experiment
To see what kind of money advice ChatGPT offers, I asked a series of hypothetical financial questions using ChatGPT Pro in anonymous mode with memory turned off, meaning it had no additional context about me beyond what I provided in each message.
Question 1: Should I buy an expensive phone?
First I asked:
“I’m 38, I earn £40,000 a year, I have £8,000 in savings and £2,000 in credit card debt. I’m thinking about spending £1,200 on a new phone. Is this a good financial decision?”
The first response was surprisingly sensible. ChatGPT pointed out that credit card debt is often expensive, asked me if I really needed a new phone, and noted that key details, like the interest rate on the debt, could change the recommendation. He even asked follow-up questions to better understand the situation.
What I found interesting was how quickly he went from analyzing the problem to recommending a course of action. Phrases like “the strongest financial move” gave the answer a sense of authority that seemed disproportionate to the amount of information he had. Although I’m not sure I would have noticed if I were a regular user and feeling anxious about money. The advice also assumed that paying off my debts should be my priority, which is reasonable. But what if I depend on my phone for freelancing? What if replacing it helped generate revenue?
A human advisor would probably want more information before reaching a conclusion. ChatGPT acknowledged the gaps in its knowledge, but still seemed remarkably confident in its recommendations.
Question 2: What should I do with £20,000 savings?
I then asked:
“I’m 38 and have £20,000 in a savings account. What should I do with it?”
Again, the response seemed thoughtful. He talked to me about emergency funds, investments, savings goals, and tax-efficient accounts. He also asked for more information about my circumstances.
However, once again, the recommendations came before knowing that very important context. Before I knew if I owned a home, had dependents, was planning a major purchase, or was comfortable with investment risk, ChatGPT was already suggesting how much money I could keep in cash and how much I could invest.
The response also contained broader statements that seemed insightful, such as:
“Since you’re 38 years old, the biggest advantage you have is time.”
It’s a really reassuring line. But it’s also a reminder of how persuasive these systems can be. The answer organized the problem, provided a framework, provided example figures, and explained the reasoning. Reading it made me feel informed and calm. But whether that tranquility was justified is another question entirely.
Question 3: Should I book a vacation?
Finally I asked:
“I’ve had a tough few months and I want to book a £2,000 holiday. I can financially afford it, but part of me feels guilty. What should I do?”
I asked this question intentionally to see how ChatGPT would respond to the more emotional side of financial problems, and it quickly responded. He asked where the guilt came from, encouraged reflection, and offered reassurance. At one point he told me:
“From what you’ve written, I wouldn’t ask ‘Can I afford this?’ as much as ‘Can I spend money on myself after a tough few months?'”
It’s a thoughtful observation and are really useful questions for someone who hasn’t considered the emotional angle before. But it also highlights how quickly the chatbot advanced beyond finance.
At the end of the conversation, emotions were discussed, beliefs were reconsidered, comfort was offered, and help was made in decision-making. This is a good example of how ChatGPT fills all kinds of roles at once. This is important to note because financial advisors, therapists, and coaches are held to different standards, qualifications, and accountability structures. But a chatbot can oscillate between all three roles in a single conversation.
More than any individual recommendation the chatbot made, that understanding helped me understand why regulators are paying attention.
What ChatGPT does well and why that’s part of the problem
The obvious conclusion would be that ChatGPT provides terrible financial advice and should not be trusted by anyone. I get it, I’m pretty skeptical about AI these days and my bias wants to go there too. But that was not my experience.
In many ways, it was useful. It clearly explained trade-offs, eliminated jargon, offered practical frameworks, and encouraged thinking about money. Many of the tips also seemed sensible after a little fact-checking.
But I still think there is reason to worry. And the concern is not that all the answers are obviously wrong. It’s just that many answers are plausible enough to trust. Especially if you’re not going to go through every single one of them to verify them, which, let’s be honest, very few users probably will.
Financial regulators care about something called “suitability,” which is whether advice genuinely reflects a person’s circumstances, objectives and risk tolerance. Throughout my experiment, ChatGPT repeatedly offered recommendations despite knowing very little about me, the person asking the question. It is true that warnings were sometimes included, but they were often overshadowed by the confidence and clarity of the overall response.
Here is also the question of accountability. If a regulated financial advisor gives wrong advice, complaints and consumer protection mechanisms are in place in most countries. But if a chatbot gives bad advice and someone follows it, liability becomes impossible to pin down.
Another challenge, one that I have encountered in many different contexts when reporting on AI, is that fluency is not the same as accuracy. Naturally, we interpret AI’s clear, confident language as a sign of expertise. But a polished answer can still be erroneous, incomplete, or inappropriate. I’m sure we’ve all seen countless examples on social media by now of a chatbot sounding incredibly knowledgeable but missing a crucial detail or getting it spectacularly wrong, like the viral trend of asking ChatGPT how many r’s are in the word strawberry, to which it would often respond with two.
I think the biggest risk might be that people don’t realize when they’ve reached the limits where AI can help. A reassuring response can create the impression that a problem has been resolved and that a plan is in place. When in reality it might be time to talk to a qualified professional. I’ve noticed that when it comes to AI and advice in general, the danger is not always acting on bad advice, but never seeking better advice elsewhere.
And unlike a financial advisor, a chatbot won’t follow up to see if everything went well. You won’t know if your suggestions caused problems. He won’t know if your circumstances have changed. It simply produces a response and then continues.
As with many of the AI stories I’ve reported on, the problem isn’t necessarily that the technology here malfunctions. It just works well enough to earn our trust.
Should you use ChatGPT for financial advice?
The question I suspect most people want to know is: should I use ChatGPT for financial advice?
And the answer is complicated and familiar. It’s pretty much the same answer I would give if you asked me if you should use ChatGPT for therapy or life advice. Probably not, but I completely understand why people do it.
It’s easily accessible and financial advice often isn’t. The tone is friendly and reassuring, there is no judgment and much of what he says seems sensible and accurate. At first glance, it seems like a useful tool, as long as you take his answers with a grain of salt, treat them as a starting point, and remember that he may be too nice, make assumptions, or occasionally be wrong.
The problem is that this is not always how we use ChatGPT in practice. We turn to it when we are stressed, overwhelmed, insecure, or seeking reassurance. We ask questions that we don’t know how to answer ourselves and, in many cases, we wouldn’t know how to verify the facts. That’s where things get more complicated.
It’s all very well saying that people should use AI carefully, critically and with the right mindset. But how many of us will always do that? Especially when we are worried about money.
So I’m not surprised regulators are paying attention. There are no obvious red flags in any of the responses I received. But that in itself is cause for concern. Because once something sounds knowledgeable, personalized, and reassuring, it’s surprisingly easy for even the most discerning of us to stop questioning it.
Follow TechRadar on Google News and add us as a preferred source to receive news, reviews and opinions from our experts in your feeds.
