Planners divided on using AI for suitability reports

Planners from three firms discuss their experimentation with generative AI and where they draw the line between human and machine work.

See Citywire 230626

The story is essentially this: advice firms are experimenting with AI, but remain split on whether it should draft suitability reports or merely support the human adviser/paraplanner.

The key divide appears to be:

Some firms are using AI for transcription, meeting notes, file checking, client-goal extraction and consistency checks. Connor Broadley, for example, says it uses Saturn for meeting assistance and fact-checking against suitability reports, but still keeps the recommendation with advisers.

Others are more cautious because suitability reports are regulated evidence documents, not just admin outputs. The concern is accuracy, data protection, hallucination, audit trail, and who is accountable if the report misrepresents the client’s position.

The FCA’s current stance is important: it does not plan separate AI rules, but expects existing frameworks to apply, including governance, SM&CR accountability, Consumer Duty, operational resilience and fair outcomes.

My take: AI suitability reports are not the real issue. The real issue is whether AI is being used to reinforce adviser-led dependency or to improve client understanding and agency.

A good boundary would be:

AI can summarise, structure, cross-check, flag gaps, compare evidence, and improve plain English.

AI should not decide suitability, manufacture rationale, obscure trade-offs, or create the appearance that the client understood something they did not.

For AoLP, this is another proof point: the profession is automating advice documentation before it has fully understood agency restoration. The next developmental step is not “faster suitability reports.” It is clearer decision support, better client comprehension, and evidence that the human being—not the machine, firm, or adviser—remains the centre of authority.