For compliance leaders, there’s no worse outcome than a vulnerable customer slipping through unnoticed. The FCA’s latest review confirms that firms meeting Consumer Duty expectations are proactive – they identify vulnerability early, encourage disclosure, and tailor support to each individual.
Increasingly, the regulator is highlighting artificial intelligence (AI) as a tool that makes this best practice possible at scale.
Why the FCA sees AI as best practice
In its multi-firm work on customer journeys, including power of attorney and bereavement cases, the FCA found that the most effective firms:
- Identify signs of vulnerability early and encourage customers to share their needs.
- Equip and empower frontline staff with the skills and clear escalation routes to respond appropriately.
- Tailor support flexibly to each customer’s situation.
- Centralise vulnerability data so customers don’t have to repeat sensitive details.
- Use AI to help identify potential characteristics of vulnerability.
Under Consumer Duty, firms must monitor and test outcomes for customers in vulnerable circumstances and have the evidence ready. For many, AI is becoming a practical way to achieve this consistently.
The challenge with traditional approaches
Even well-intentioned firms face common challenges:
- Limited sampling: Reviewing 2–5% of calls leaves the majority unchecked.
- Subjective interpretation: Different reviewers can reach different conclusions on the same call.
- Delayed identification: Vulnerability may only be spotted days or weeks later.
- Missed subtle cues: Hesitation, tone changes, or indirect hints often slip by.
The FCA’s examples of good practice – centralised records, early identification, and staff empowerment – are difficult to achieve at scale without technology.
AI: From FCA guidance to operational reality
By naming AI in its examples of good practice, the FCA is recognising its ability to:
- Monitor 100% of interactions
- Detect linguistic, tonal, and contextual cues: Not just obvious keywords.
- Reduce inconsistency: Using trained models to flag potential vulnerability in a uniform way.
- Centralise and share data: Automatically storing indicators so frontline and specialist teams have the same view.
FCA best practice → AI capability mapping:
FCA best practice | How AI supports it |
|---|---|
Identify signs of vulnerability
| Analyses all interactions for subtle language, tone, and pattern-based cues |
Encourage disclosure | Flags moments for agents to ask supportive, open questions
|
Empower frontline staff | Provides consistent alerts to guide next steps |
Centralise vulnerability data | Logs indicators in a central system, accessible across teams |
Better outcomes, stronger evidence
AI-assisted detection supports Consumer Duty compliance by:
- Flagging risk earlier: Giving frontline staff more time to act.
- Freeing QA time: Allowing human reviewers to focus on coaching and intervention.
- Providing a defensible audit trail: Showing exactly when and how vulnerability was identified.
- Supporting escalation: Ensuring specialist teams receive timely, relevant information.
Overcoming concerns
- “AI will replace human judgement” → In reality, AI surfaces risks and humans make the final decision.
- Data privacy → Leading solutions comply with GDPR and FCA data handling standards.
- Integration → Many AI tools connect directly to existing telephony and QA systems with minimal disruption.
Getting started the FCA-aligned way
You don’t have to transform your operations overnight. Start with:
- Run a targeted pilot: Focus on a high-priority customer journey (e.g., bereavement) and compare AI detection rates with current QA.
- Train on your data: Fine-tune detection models to reflect your products, customers, and vulnerability profiles.
- Embed governance: Define when AI alerts trigger human review, and how those outcomes are recorded.
These steps align directly with the FCA’s own good practice recommendations – making AI adoption both compliant and practical.
The bottom line
The FCA now recognises AI as a best-practice tool for identifying customer vulnerability. That’s not just a nod to innovation, it’s a clear signal that proactive, data-driven detection is becoming the new standard of care.
For compliance leaders, the choice is simple: continue relying on sampling and subjective judgement, or adopt tools that can scale detection, support your teams, and provide the evidence the FCA expects.
