FCA authorised · FRN 724952 0117 325 0027Quote & buy →
Aipedia: Claims management companies
CMCs can use AI voice agents, but the same marketing and contact restrictions apply as to human callers, and the firm must keep records of what was said. Automated calls that pressure or mislead consumers quickly become complaints and regulatory questions.
Part of: AI in Claims management companies
In short
AI voice agents now call back enquiries, qualify leads and handle inbound claimant calls. Claims management faces tight restrictions on unsolicited marketing calls, and AI does not change what is permitted. SYSC 9.1.1R means recordings, transcripts and prompt versions should be kept, and the FCA expects firms to offer vulnerable customers a non-AI route. Problems from automated contact mostly lead to complaints, ICO action and FCA scrutiny, which PI does not usually cover. PI may respond where a bot's wrong information causes a consumer loss, subject to the wording.
AI voice agents can hold natural-sounding conversations, so claims firms are using them to call back web enquiries, qualify leads, collect claim details, chase documents and answer inbound calls out of hours. Some systems transfer to a person when the conversation gets complicated; others run whole calls end to end. For a business that lives on contact rates, the appeal is obvious. The risk is that a script followed at scale can mislead, pressure or simply fail many consumers before anyone listens to a recording.
Unsolicited marketing calls are restricted under electronic marketing law enforced by the ICO (see our PECR entry), as well as FCA conduct rules on communications. Check the current rules with your compliance adviser before running any outbound AI calling. Using AI does not change what is permitted: an automated outbound call is still a call made by the firm, and recorded or automated calls can face stricter consent requirements than live ones.
Illustrative scenario. A CMC deploys an AI agent to call people who filled in a web form months earlier. The prompt tells it to "secure agreement" before ending the call. It keeps going when an elderly caller says she is confused, and tells her there is no cost unless the claim succeeds without mentioning termination fees. Her family complains, the complaint reaches the Ombudsman, and the FCA asks for recordings of a sample of calls. The firm finds it cannot produce transcripts for the prompt version used that month.
Voice agents can also invent answers. Asked whether a claim is likely to succeed or how much it might be worth, a generative model may give a confident figure that no one at the firm would stand behind. Because the agent is usually supplied by a vendor, the firm may not even see the change when the underlying model is updated. The FCA is clear that firms remain responsible for outsourced AI services, so vendor updates need testing like any other change.
Most problems with automated contact lead to complaints, regulatory scrutiny and ICO enforcement rather than negligence claims. Fines, regulatory penalties and refunds of fees are commonly outside PI cover, and some policies exclude claims arising from marketing activity. Where a bot gives a consumer wrong information about their claim and they suffer a loss as a result, a PI policy may respond if held, subject to the wording. See Does claims firm PI cover AI errors? and our guide to AI chatbot insurance.
Apex arranges PI for claims management companies and can explain how your policy treats automated customer contact and the complaints that follow. See AI in claims management and our page on vulnerable customers.
Only within the marketing and contact rules that apply to claims management, which are strict for unsolicited calls. AI does not change what is permitted. Check consent and current rules with your compliance adviser before any campaign, and keep full records.
SYSC 9.1.1R requires records sufficient for the FCA to monitor compliance. For AI calls, recordings or transcripts plus the prompt version used are the practical way to show what consumers were told, and the FCA's good-practice page expects auditable records.
It is good practice. The FCA's good-practice page encourages telling consumers how AI is used and what its limits are, and a clear statement at the start of the call reduces the risk of a consumer feeling misled. Offer a quick route to a person as well.
Usually not. Regulatory fines are commonly excluded from PI and may be uninsurable. PI is aimed at civil claims for negligent services, so it may respond if a bot's wrong information causes a consumer a loss, subject to the wording.
Tell us about your firm and the work you do. A broker will review how your current PI treats AI-assisted work, or arrange new cover. No fee for a quote. Or call 0117 325 0027.
Start a PI quote Request a callbackApex Insurance Brokers Limited is authorised and regulated by the Financial Conduct Authority. Registered in England and Wales, company number 07014570. This page is general information about how AI is used in the profession and how insurance may respond. It is not legal, regulatory or insurance advice on your circumstances, and it does not guarantee that cover will be available or on what terms.
