The FCA's Consumer Duty applies to AI-driven customer interactions just as it applies to human-driven ones. The four outcomes, products and services that meet customer needs, fair value, customer understanding, customer support, each have AI implications that FCA-regulated firms need to operationalise.
This article walks through what each Consumer Duty outcome requires when AI is part of how the firm delivers products, services, and support to customers.
Products and services that meet customer needs
Consumer Duty's first outcome is that products and services should be designed to meet customer needs. AI affects this outcome through product recommendation engines, customer segmentation, AI-assisted product design, and the targeting decisions that determine which customers see which products.
Operational implications
● AI-driven product recommendation must surface products genuinely fitted to the customer's needs, not products optimised for the firm's commercial outcomes at the expense of fit
● Customer segmentation that drives differential product offering needs to be reviewed for fairness and for alignment with customer needs
● AI-assisted product design and parameter setting should reflect customer outcomes rather than only firm metrics
● Target market definitions need to account for AI-driven targeting decisions, including the customers AI excludes as well as the customers it includes
● Vulnerable customer identification through AI must be carefully designed, AI may both improve identification and introduce its own biases that affect specific vulnerability dimensions
Common pitfalls
AI recommendation systems trained primarily on commercial outcomes (conversion rate, revenue per customer) without sufficient weight on customer fit; segmentation that produces differential treatment without clear customer-needs justification; target market definitions that AI implementation effectively narrows or expands beyond what the documented target market specifies.
Fair value
The second outcome is that products and services should provide fair value, the relationship between price paid and benefits received should be reasonable. AI affects fair value through AI-driven pricing, AI-assisted commercial decisions, and the broader value chain shaped by AI.
Operational implications
● AI-driven pricing must support fair value outcomes, not exploit information asymmetries, behavioural biases, or differential willingness to pay in ways that produce unfair value
● Price walking through AI is a specific FCA concern, AI that systematically increases pricing for customers who do not switch needs review against fair value standards
● AI-assisted underwriting and risk-pricing should reflect actuarial reality, not produce systematic differential pricing that does not relate to risk
● AI-driven optimisation of commercial outcomes needs to be balanced against customer outcome metrics; pure commercial optimisation risks Consumer Duty failures
● Documentation of fair value assessments must reach AI-driven components of pricing and commercial decisions, not only manual processes
Customer understanding
The third outcome is that communications should support customer understanding. AI affects customer understanding through AI-generated communications, chatbots, AI-assisted advisers, and AI-driven personalisation of information.
Operational implications
● AI-generated customer communications must be designed to support understanding, not just technical compliance, readability, clarity, completeness, appropriateness for the customer
● Chatbots and AI agents serving customers need to produce content that customers can act on, with appropriate escalation when AI cannot deliver the needed clarity
● AI-driven personalisation that adjusts information to different customers must be reviewed for whether the personalisation supports understanding or fragments it inappropriately
● Vulnerable customers may need specific accommodation, AI communications optimised for the average customer may fail vulnerable customers in ways that breach Consumer Duty
● Documentation of customer understanding testing must include AI-driven communications, not only manually drafted material
Customer support
The fourth outcome is that customer support should help customers realise the benefits of the products and services they buy. AI affects customer support through chatbots, AI-assisted human agents, automated triage, and AI-driven self-service.
Operational implications
● AI-driven customer support must deliver good outcomes, including for vulnerable customers, with effective escalation when AI is not the right tool
● Customer support that fails to support good outcomes, where AI cannot resolve the issue, where AI handles vulnerable customers inappropriately, where AI provides incorrect information, is a Consumer Duty issue regardless of cost efficiency
● Escalation pathways from AI to human support need to be accessible and effective, buried escalation that customers struggle to find is itself a Consumer Duty issue
● Monitoring of AI support outcomes must include the right metrics, not just resolution rate but resolution quality, customer outcome, complaint patterns, and vulnerable customer experience
Vulnerable customers and AI
Vulnerable customers require particular attention under Consumer Duty. AI implications for vulnerable customer support deserve specific operating discipline.
● AI vulnerable customer identification can improve consistency and coverage, but introduces its own biases, including failing to identify vulnerability patterns the AI was not trained to recognise
● AI-driven interactions with identified vulnerable customers must adjust appropriately, not simply apply the standard AI interaction with vulnerability flagged in the background
● Escalation to specialist support for vulnerable customers needs to be reliable when AI identifies the need
● Testing of AI interactions with vulnerable customers needs to use vulnerable-customer-relevant test cases, not only general customer test cases
● Outcome monitoring needs to specifically include vulnerable customer outcomes, not aggregate metrics that obscure vulnerable customer experience
Documentation and evidence
Consumer Duty operates on evidence. AI-related Consumer Duty obligations require specific documentation:
● AI inventory mapped to customer journeys and to Consumer Duty outcomes
● Outcome testing methodology for AI-driven elements of customer journeys
● Monitoring metrics covering AI-driven customer outcomes, not only operational AI metrics
● Management information for senior management oversight of AI Consumer Duty performance
● Complaint analysis specifically covering AI-related customer issues
● Improvement actions documented and tracked
The shift to make
Stop treating Consumer Duty as a customer journey exercise that AI happens to be part of.
Start treating it as the customer outcome framework that AI must specifically deliver against, with AI-specific operational discipline, AI-specific testing, AI-specific outcome monitoring, and AI-specific senior management oversight.
Firms operating this way pass FCA Consumer Duty engagement on AI as a matter of course. Firms operating AI as a separate workstream from Consumer Duty discover, often during supervisory dialogue or complaint patterns, that the integration gap is exactly what the FCA looks for.
