FCA authorised · FRN 724952 0117 325 0027Quote & buy →
Aipedia: Dentists
Dental practices can use AI to suggest NHS treatment bands and complete claims, but the performer remains responsible for every claim submitted. A coding error repeated across hundreds of courses can look like dishonesty, and probity matters are not handled like clinical negligence claims.
Part of: AI in Dentists
In short
Some practice systems can read clinical notes and suggest the NHS treatment band and claim details, or complete claims automatically. If the logic is wrong, or the notes do not support the band, the practice can overclaim at scale. GDC standard 1.3 requires dental professionals to be honest and act with integrity, and NHS claims carry probity expectations from commissioners. A probity investigation is different from a negligence claim: there is often no patient harm, the money at stake is NHS funding, and the regulatory consequences can be severe. Clinical indemnity usually responds to patient claims, not repayment of overpaid NHS funds.
In England, NHS dental care is claimed as courses of treatment in bands, and contract performance is measured in units of dental activity. Completing claims correctly takes time and attention. Practice management systems may read the clinical notes and charting, suggest the band and claim details, flag missing information and, in some set-ups, submit claims in batches. Some tools also suggest additional items that the notes appear to support.
Used well, this reduces underclaiming and admin errors. Used badly, it can multiply one mistaken assumption across a whole list.
There is an important difference between a tool that checks claims a person has prepared and one that prepares and submits them. A checking tool adds a safeguard. A submitting tool removes the moment where someone who knows the patient looks at the claim, which is often where errors used to be caught. Practices should decide deliberately which model they are using and who signs off.
The GDC has no AI-specific rule here, but automated claiming does not change who is responsible for each claim.
Illustrative scenario: a practice switches on automated claim completion. A template used for urgent appointments includes a charting entry that the tool reads as a filling, so many urgent courses are submitted at a higher band than delivered. Months later, a routine review by the commissioner identifies an unusual claiming pattern. The performers whose numbers were on the claims are asked to explain, and the commissioner seeks repayment.
Other issues include splitting courses of treatment, claiming items that were planned but not completed, and AI suggestions to add items the notes only loosely support.
These are different problems. A clinical negligence claim alleges that a patient was harmed by care below a reasonable standard, and compulsory indemnity exists so the patient can be compensated. A probity investigation asks whether claims for public money were accurate and honest. There may be no harmed patient, the consequences can include repayment, contract action and referral to the GDC, and dishonesty findings can end a career. Defence organisations may offer discretionary help with regulatory and contractual matters, but repaying overclaimed NHS funds is not usually something indemnity pays. One defence organisation's published position on AI excludes software defects. See does PI cover AI mistakes for the wider picture.
Related: AI treatment planning, AI scribes and notes, dental indemnity and AI, the GDC position and the dentistry AI hub.
Apex arranges indemnity for dentists. A broker reviews how your arrangements treat regulatory and probity matters, as well as clinical claims, where AI is used in practice systems.
Usually, yes. The performer and contract holder remain responsible for the accuracy of claims submitted in their name. A software error may explain how it happened, and may support an argument that there was no dishonesty, but it does not transfer responsibility to the supplier.
Not usually. Indemnity is designed to compensate patients harmed by clinical negligence. Repayment of NHS funds that should not have been claimed is not normally treated as an insured loss, although defence organisations may offer discretionary help with the investigation itself.
Keep evidence of checks: who approved claims, how the tool was tested, when you found the error and what you did about it. Prompt correction and reporting is far easier to defend than a pattern that continued after warning signs appeared.
Generally, yes. A tool that flags possible errors for a person to review adds a safeguard, while automatic submission removes the human check. Whichever model you use, the performer remains responsible for each claim, so keep a record of who reviewed what.
Tell us about your practice and the work you do. A broker will review how your current cover treats AI-assisted work, or arrange new cover. No fee for a quote. Or call 0117 325 0027.
Start an indemnity 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.
