In this episode, leading business journalist Ruth Sunderland sits down with Continuum’s Managing Director, Martin Brown—and an AI named Claude—to discuss the rapidly evolving role of AI in financial advice.
The reality of AI in financial advice
Ruth: It is time you and I talk about AI.
Martin: Oh dear. I will just say it straight away, you’ve picked a subject which is not my specialism. This is going to be an interesting car crash of an interview.
RS: Now we’re both laughing.
MB: We are.
RS: But tell me, where have you got AI in the business and where would you not countenance it?
MB: We’ve been very careful about implementing AI simply because of the lack of accuracy it currently presents. We’ve been inundated with calls from providers that claim to be 85% accurate, 88% accurate or 90% accurate. But for us, being wrong 15% of the time is not acceptable so we’re quite nervous about it.
RS: How are you addressing that?
MB: We have a steering group and we set up a project to look into it in some detail. As you can tell, Ruth, I’m not the expert and probably not the perfect one to be commenting on AI.
RS: No, you’re the perfect person because none of us yet are experts in AI. We are all learning and it’s changing so quickly. You are running a real business in real time and you have to deal with it here and now. So tell me, are you drawing a line between AI in the back office and AI being used in direct engagement with clients? Is that an important distinction for you?
MB: Yes it is. AI can bring huge benefits in adding efficiencies behind the adviser to streamline the behind-the-scenes work, for example, diary management. We can use it to support the running of the business. AI can go through all the data and produce a streamlined summary report on performance in different areas, which we can follow up with human intervention. It can save hours of deep dive analysis which is a huge advantage. But when it comes to clients, we want human interaction. You can’t replace that.
Why emotional intelligence beats algorithms
RS: So much of financial advice is not just about facts and figures but about emotions. Does that mean AI can never replace a human adviser?
MB: An adviser brings emotional intelligence and is aware of the context, the way things are being said, the expression on someone’s face or the tone of their voice. The emotional side of advice is one of its greatest attributes. What happens is you open up a partnership with an adviser, it enables you to think about what you want, how you want to live, and to understand what is meaningful to you. And until you open up that conversation, you don’t know how important it is. A bot does not have empathy.
“An adviser brings emotional intelligence and is aware of the context, the way things are being said, the expression on someone’s face or the tone of their voice. The emotional side of advice is one of its greatest attributes.”
Martin Brown
RS: Do you think AI could risk opening up more vulnerability to financial advice scams and fraud?
MB: Yes, it is already a serious epidemic. You might be being impersonated by a bot right now.
RS: I’m not. It’s me, honestly.
MB: OK I believe you. But I was watching a programme six months ago about AI and this lady was presenting and then at the very end she revealed that she had been an AI avatar all along. She disappeared and that was the end. You just get sucked in.
RS: Moving on, if advisers were to outsource too much of their work to AI, do they risk losing the very thing that makes the job so satisfying, which is their relationship with the client?
MB: Absolutely. Which is why I think AI should be used in the back office systems to enable them to spend more time with the client. Because if you speak to advisers, what do they love? Helping clients, providing solutions, moving them forward in life and changing things for the better. What they don’t love is being bogged down with administration and all the painful things that go on behind the scenes.
This is my key point from the steering group. People were saying: I’ve seen this bit of AI and it does this, and I’ve seen that AI and it does that.
I said: “Stop for a moment. Let’s just think about what we want AI to do. The goal is caring for the client and the best possible advice for the client. So let’s make sure we spend more time on that. Don’t just chase a shiny toy.”
Some of the things people were suggesting simply weren’t relevant to us at all. It’s like someone said a while ago they had AI that can translate everything into Japanese. But we don’t have Japanese clients. My view is: if it’s relevant and it helps then use it. If it doesn’t, wait. Some of the most profitable decisions you make the ones you walk away from.
My view is: if it’s relevant and it helps then use it. If it doesn’t, wait. Some of the most profitable decisions you make the ones you walk away from.
Martin Brown
Claude the AI enters the interview
RS: Now Martin brace yourself – because Claude, the AI, has asked me to put some questions to you.
MB: Oh what are you doing to me now?
CLAUDE: Martin, if I showed up at Continuum as a job applicant, I would be qualified on paper. I would be tireless. I would be able to process thousands of documents in seconds. I never go off sick and I’m never in a bad mood. But I have a tendency to state things with total confidence that turn out to be wrong, including financial calculations and regulatory rules. Would I even get through your door?
“I have a tendency to state things with total confidence that turn out to be wrong, including financial calculations and regulatory rules. Would I even get through your door?”
Claude (The AI)
MB: I respect the honesty. We look for people who are full of honest endeavour, but financial services is a people business.
CLAUDE: I’m taking that as a no, which is what I suspected. But you have human advisors who make mistakes. Also, values could be trained into me. Will you give me a second chance?
MB: Hold on. I didn’t say I wouldn’t interview you. So there you go, taking it incorrectly. I am laughing because that’s a classic example. You could programme in values but you can’t create a real personality or a natural ability to get on with people.
Transparency and trust in wealth management
CLAUDE: You talk about trust which I find it genuinely difficult because I know I can’t be trusted unconditionally. I make mistakes and I don’t always know when I’m making them. Do you think that your clients should be told when I have been involved, even if it’s only in the back office?
MB: It depends on the role you are playing, Claude. If there’s any human interaction then we would make people aware. But if it’s purely in the back office, then there would be no need.
RS: Claude says thank you for answering his questions even if you didn’t hire him.
MB: I’d invite him around for a dinner party though because I think the conversation would be great fun.
RS: He is very honest but he’s a bit one-dimensional and he can’t provide emotional security like a human adviser.
MB: Absolutely not. But an even cleverer Claude might be able to transform life significantly. I think AI should be used to do something great, to change the world in a meaningful way.
Key takeaways: Navigating AI in financial advice
- Back-office automation over client replacement: The consensus is that technology excels at backend administrative tasks, but human intervention remains non-negotiable for client-facing relationships.
- The empathy gap: True wealth management relies heavily on context, tone, and emotional intelligence—areas where a digital bot simply cannot compete with a human professional.
- A measured approach to tech: Successfully adopting AI in financial advice means walking away from “shiny toys” and only implementing tools that directly improve accuracy and client care.
