GDC's consent standard is binding. BDA's AI guidance isn't.

Ask a dental principal whether a patient needs to be told before an AI charting tool or diagnostic aid is used chairside, and most will say yes, then struggle to name the rule that actually says so. Two different bodies have something to say about it: one whose standard is legally enforceable and does not mention AI at all, and one whose AI-specific guidance is detailed and useful but carries no statutory weight on its own. If you are weighing up an AI tool ahead of opening a third site, this is the split worth getting straight before you sign anything.

It matters more than it looks. A confused answer to "did the patient know AI was involved" is not just an awkward moment in an inspection. It is the kind of gap a fitness-to-practise complaint or a civil claim can be built around. Knowing which document actually binds you, and which one is advisory, is a five-minute fix that closes that gap.

The one consent standard that's actually binding

The General Dental Council regulates every dental professional in the UK by statute and keeps the register that determines who is legally permitted to practise (General Dental Council, Standards for the Dental Team). Principle Three of that document, "Obtain valid consent," sets out Standard 3.1: "You must obtain valid consent before starting treatment, explaining all the relevant options and the possible costs." A supporting note adds that a signature on a form matters, but it is the discussion with the patient that actually determines whether consent is valid.

That is the whole of the GDC's binding consent requirement. It is not AI guidance. It has applied to every registrant regardless of what tools are used in the surgery, and it is enforceable: the GDC can and does act on fitness-to-practise concerns where consent was not genuinely obtained, whatever technology was or was not involved in delivering the treatment.

So what for you: if a complaint ever asks whether a patient consented to AI-assisted diagnosis or record-keeping, Standard 3.1, not any AI-specific document, is the rule your answer gets measured against.

What GDC's own standards don't say about AI

Read Principle Three in full on the GDC's own Standards for the Dental Team site and AI does not appear once. The GDC has engaged with the subject elsewhere: in August 2025 it published a Rapid Evidence Assessment on artificial intelligence and dental service provision, carried out by the Peninsula Dental School at the University of Plymouth, reviewing 45 international studies published since 2020 (General Dental Council, Artificial Intelligence and Dental Service Provision: A Rapid Evidence Assessment, August 2025). The report found promising applications in implant robotics, caries detection and remote monitoring, but also flagged that it found little evidence on ethics and data protection questions specific to AI, and called for UK-specific research to fill that gap.

That is a research review, not a new consent standard. As of this writing, there is no GDC-specific clinical guidance interpreting how Standard 3.1 applies to AI tools used in daily practice.

So what for you: do not wait for the GDC to publish an AI-specific consent rule before acting. Standard 3.1's general wording, "explaining all the relevant options," already covers an AI-assisted option. The obligation exists now, whether or not the word "AI" appears anywhere in the GDC's text.

BDA's AI guidance says more, but it isn't the law

The British Dental Association is a professional membership body and trade association for dentists, not the statutory regulator. That distinction matters here specifically, because BDA membership carries real professional and practical weight, but not the GDC's enforcement route.

In May 2026 the BDA added a new advice page, "Using artificial intelligence (AI) in dental practice," covering regulatory expectations, data protection risk, and the limits of relying on AI-generated advice for clinical or business decisions (British Dental Association, advice page, last updated May 2026). Expert members can also download a template AI Use and Governance Policy. A companion blog from the BDA's Acting Head of Operations in Advisory Services argues that AI is a useful adjunct for summarising information and admin, comparable to how loupes enhance a dentist's vision without seeing for them, but that professional accountability for any decision still sits with the dentist, not the software (British Dental Association, "Should dental practices use AI?", 19 May 2026).

This is more specific and more practical than anything the GDC has published on AI. It is also, on its own, non-statutory. A dentist who ignores the BDA's guidance faces no direct GDC sanction for that alone. Where it bites is indirectly: if the AI disclosure a practice gives a patient is vague or generic enough that a regulator would not call it genuinely informed, that failure lands back on Standard 3.1, not on the BDA guidance itself.

So what for you: treat GDC's Standard 3.1 as the reason consent is compulsory, and the BDA's May 2026 guidance, including its Expert-tier governance policy template, as a working answer to what your practice's AI disclosure and internal policy should actually contain.

Consent for AI isn't a one-off signature

Standard 3.3 of the GDC's consent principle states that "you must make sure that the patient's consent remains valid at each stage of investigation or treatment," and describes consent as an ongoing process rather than a single event. Neither the GDC nor the BDA has published a document that says explicitly "adding a new AI tool counts as a stage requiring fresh consent." That is our own reasonable reading of Standard 3.3 applied to a new situation, not a quoted regulatory rule, and it should be treated as a defensible interpretation rather than a settled position.

It is a defensible one, though. Moving an existing patient from handwritten charting to an AI-assisted charting or scribe tool, or adding AI-based image analysis to routine radiograph review, changes how their information is captured and assessed in a way that a patient who consented years ago was never actually told about.

So what for you: build a checkpoint into your onboarding process for any new AI tool, so existing patients are told and given a genuine chance to ask questions, rather than assuming a signature taken before the tool existed still covers it.

A scope caveat worth stating plainly

GDC's Standard 3.1 binds you as an individual registrant, wherever you practise. It is a separate compliance track from CQC's oversight of the practice as a provider organisation in England, which folds AI-specific governance checks into its well-led framework rather than into consent (see our earlier analysis of CQC's 2026 well-led rewrite for dental practices). Adding a third site does not multiply your GDC consent obligation, since it attaches to each registrant personally, but it does multiply the number of CQC-registered locations whose AI governance paperwork needs to be consistent. Treat the two as parallel requirements, not one covering the other.

So what for you: if you have reviewed consent but not your CQC well-led AI documentation, or the other way round, you have only closed half the gap.

What this means for your practice

Four actions, in order of priority.

First: write a specific AI disclosure for each AI tool in use, covering its purpose, what happens to patient data, and its main benefits and risks, in plain English rather than a vendor's marketing copy. This is the content the BDA's May 2026 guidance calls for, and the evidence the GDC would look for under Standard 3.1.

Second: treat the introduction of any new AI tool, or a material change to an existing one, as a consent-review trigger for current patients, not something covered only by new-patient intake paperwork.

Third: document the consent conversation itself, not just a signed checkbox, so you can evidence it was genuinely informed if a complaint or review ever asks. For a multi-site group, this needs a shared template applied consistently across every location and every associate, rather than left to individual discretion, a discipline our earlier analysis of multi-site dental AI standardisation covers in more depth.

Fourth: keep GDC consent compliance and CQC well-led AI governance as two tracked, separate workstreams rather than assuming one satisfies the other, particularly as you open a third site and the number of CQC-registered locations grows. If you want an independent view of where your practice's consent and governance paperwork actually stands against both, the AI Opportunity & Growth Assessment™ reviews both as part of its standard scope, and you can book a 20-minute call to talk through where your group sits before committing to anything further.

The single most important thing to take from this

Standard 3.1 was written before generative AI existed in its current form, and it does not need updating to apply to it. It already covers any treatment you provide, AI-assisted or not, and it is the standard a regulator will actually use if your consent process is ever challenged. The BDA's May 2026 guidance is not a substitute for that standard. It is the most useful available answer to what a compliant AI disclosure and governance policy should say. Practices that treat "we followed the BDA's advice" as the same thing as "we met the GDC's binding consent requirement" are the ones most likely to discover, at the worst possible moment, that a form signed years ago never covered the tool being used today.