FT Logo

Copy article

FT Editor Banner

How are advice firms addressing the ethical considerations of AI?

Journalist: Sonia Rach

ended 19. January 2026

AI is having an increasing role in financial planning, with many using it for completing repetitive tasks.

But how are firms addressing the ethical considerations and ensuring transparency as AI takes on more complex advisory tasks? What are some the measures in place especially when it comes to automating tasks?

I'd love to hear from you!

5 responses from the Newspage community

Copy all

Copy

Editor
Scott Gallacher
Director at Rowley Turton
At this stage, I don’t see a genuine ethical dilemma. In most firms, AI is used for analysis, research and report drafting — not to give advice. While it may inform the decision-making process, responsibility for the recommendation must always sit with the adviser.

In practice, there are three outcomes. The AI supports the adviser’s recommendation, reaches a different conclusion which the adviser understands and overrules, or prompts a rethink where there are clear, defensible reasons to change course.

The only real concern would be a fully automated advice process with no meaningful adviser oversight. The ethical risk isn’t using AI — it’s allowing it to replace professional judgement.
Copy

Editor
Nouran Moustafa
Practice Principal & IFA at Roxton Wealth
AI will change financial advice, but it must never replace it. Advice is not a calculation, it is a human, behavioural discipline. No system understands fear, hesitation, or the stories people tell themselves about money. That is where real advice lives.

Because once advice is outsourced to machines, you no longer have advisers you have software with a call centre attached.

AI should remove friction, not responsibility. Its real value is in oversight and protection, a human compliance officer can review some cases, a system can review all of them, exposing patterns, conflicts, and advice that looks right on paper but fails the client in reality.

This way we can use AI in a pro regulation in a way that can help protect vulnerable clients.

The proper role of technology is not to replace the profession, but to elevate it: to raise standards, sharpen accountability, and leave human beings to do the one thing machines never will, exercise judgement in the service of other human beings.
Copy

Colette Mason
AI Ethics Consultant at Clever Clogs AI
Most firms are using AI for basic market analysis, draft email generation, pattern spotting. That's legitimate. Logic-based automation like "If client income drops 10%, trigger adviser review"? Brilliant.

The moment AI starts recommending portfolio changes, tax strategies, or pension drawdowns without a qualified human checking recommendations, you're in trouble.

The regulatory gap is glaring. The FCA states firms must explain their decisions, but most can't tell you how their AI systems work, and are extremely reluctant to tell clients where AI is being used legitimately and responsibly.

AI deciding "this client should shift to higher-risk assets based on generic patterns" without human oversight? That's a fine and lawsuit waiting to happen.

Ethical AI means humans own all client-affecting decisions, AI provides the research and automates logic-based tasks like meeting reminders. You need to trace every recommendation back to auditable logic, and with AI that's impossible.
Copy

Rohit Parmar-Mistry
Founder at Pattrn Data
The real ethical risk isn't AI going rogue; it's firms hiding behind 'black box' algorithms they don't understand. FCA Consumer Duty demands you explain why a recommendation is suitable. If your AI generates a strategy but can't explain the logic in plain English, you aren't innovating, you’re gambling with people's livelihoods. Clients pay for expert judgment, not black-box calculation.

We must be ruthless. AI is phenomenal at the admin drudgery. That is where the tech belongs. Use it to strip away bureaucracy so advisers can actually look clients in the eye.

But letting an algorithm interpret retirement anxiety or complex family dynamics is a failure. Financial planning requires empathy and nuance, code cannot simulate that. Technology must serve the relationship, not automate it away.

Ethical AI isn't about replacing the adviser; it's about building the ultimate paraplanner. Use bots for the data, so humans can handle the trust
Copy

Patricia McGirr
Founder at Repossession Rescue Network
AI in financial planning is not a productivity story. It is a trust story, and trust is fragile. Firms say AI is only handling admin, modelling and paperwork, but clients rarely see where automation ends and judgement begins. That opacity is the risk. When systems start shaping recommendations, prioritising options or nudging outcomes, responsibility still sits with the firm, not the software. The ethical line is simple. AI can assist, but it must never decide. Every automated output needs human sign off, a clear audit trail and plain English disclosure to the client. If an adviser cannot explain how a result was produced, it should not be used. Efficiency is not innovation if it weakens duty of care. Get that wrong, and confidence collapses fast.