Advice and AI
AI can write the review. It cannot carry the responsibility, or the relationship.
The profession is being told to automate. The regulator has been clear it will not write new rules for AI, which means the old ones still apply in full. When a machine produces a client review, a named person in your firm still owns the outcome, and the client still wants a person who knows them.
The regulatory reality
What the FCA actually said
The FCA has confirmed it will not introduce AI-specific rules. Consumer Duty, the Senior Managers and Certification Regime and existing systems and controls already cover it. That is not a relaxation. It places the accountability for every outcome on a named individual, whether a human or an AI produced the work.
Read it the other way and it is a warning. A firm that automates the reviews for the clients it can least afford to spend time on, then signs its name under outputs no human checked, has not removed risk. It has concentrated it.
The human premium
Trust is the part a machine cannot reach
Advice has never really been about selecting a product. It is about trust, built over years of a client feeling genuinely understood. A person remembers the story behind the money: the family, the worries, the plans that changed, and where the client is on their own journey.
AI can summarise a fact find. It cannot know a person, hold their confidence, or sit with them through a decision that matters. For an ageing client base hitting real life events, that relationship is not a nice-to-have. It is the value, and it is the one thing automation cannot replace.
Use AI for the lifting. Keep the human for the judgement.
What AI handles well
- Drafting notes from a meeting
- Summarising a fact find
- Producing a first-pass review document
- Flagging routine checks
What only a human brings
- Trust earned over time
- Deep understanding of the client’s life
- Judgement on vulnerability and the advice boundary
- Standing behind the outcome
Where Pillar sits
The human layer, kept alive at scale
Pillar is the human layer. A trained Client Review Manager runs the review for the part of your book that costs more to serve than it generates, inside your systems and under your brand, building a real relationship with each client over time. The technology does the lifting. The human carries the trust, the judgement, the vulnerability call, the escalation and the audit trail.
You get the efficiency of automation without handing your client relationship, or your accountability, to a piece of software.