Should doctors use AI clones?

A week ago I wrote about why patients are choosing you, not your clinic; why a clinician's personal voice has become the actual trust asset; and why the practice brand should ride on it rather than compete with it. Also, Why Content is King in dental marketing.

AI clones can help you increase the volume of content without putting in hours, but where do you draw the line?

This week's question follows directly from that one, and it's one I'm being asked more often by dentists exploring the AI tools now available to them.

If your face, your voice, and your visible judgment are what's actually earning trust — what happens when an AI-generated version of you starts doing some of that talking instead?

AI avatars and voice clones can now produce a video of you explaining a procedure, answering a common question, or narrating a patient education series without you personally sitting in front of a camera. The technology is genuinely good. Good enough that the real question isn't whether it can pass as you — it often can. The real question is whether it should.

An AI clone can say your words. It can't earn what your face has already earned.

Here's the tension worth sitting with honestly: everything I've argued in this series so far — that patients trust a specific person's specific judgment, that this trust is the one asset AI-generated content can't replicate at scale — runs directly into a tool that lets you generate exactly that specific-seeming content without you actually being present for it.

Used carelessly, an AI clone is the fastest way to hollow out the very trust asset it's impersonating. Used deliberately, in the right places, it's a legitimate way to extend a clinician's reach without pretending to be something it isn't.

The difference isn't the technology. It's where you use it, and whether patients know what they're watching.

Where an AI clone genuinely helps.

Scaling repetitive patient education, not relationship-building.

Some content doesn't need your live presence to carry the same trust; it needs your accuracy, your explanation style, and your approval. A video answering "what should I eat after a tooth extraction" or "how do I care for my retainer in the first week" is information delivery, not relationship-building. An AI-narrated version, clearly produced from your actual guidance and reviewed by you, can reach far more patients than you could by filming the same explanation a hundred times individually. You can use it in aftercare content to reach out to patients and show that you actually care about them. Patients also understand that it is not possible for a dentist to reach out to every single patient, but sending a message still shows that they are important.

Multilingual and accessibility reach.

A voice clone that can deliver the same patient education in a second or third language, accurately reviewed, not just auto-translated — genuinely extends care to patients who'd otherwise get a worse, generic version of the same information, or none at all. This is one of the strongest legitimate cases for the technology in a clinical setting.

Consistency in high-volume onboarding content.

New patient welcome sequences, appointment prep instructions, post-op care reminders — content patients need reliably, on schedule, regardless of how busy the clinic is that week. An AI-delivered version of this, clearly framed as an automated system speaking on the practice's behalf, keeps the information flowing without demanding a clinician's live time for content that was never really about relationship-building in the first place.

Where it actively damages the thing it's trying to build.

The content that builds personal trust, a hard-case story, an honest answer to an embarrassing question, an opinion on a controversial treatment choice, anything a patient is watching specifically because they want to know what you think — loses its entire value the moment it's generated rather than genuinely yours. If a patient later discovers that the "you" they felt they knew was AI-generated, the damage isn't limited to that one video. It retroactively undermines every piece of real content they'd trusted before it.

Undisclosed use, anywhere.

This is the line that matters most, clinically and ethically, not just strategically. A patient who believes they're hearing directly from their dentist — about a treatment recommendation, a risk explanation, anything that touches an actual clinical decision — deserves to know whether that voice was AI-generated. Undisclosed use isn't just a trust risk. In a patient education or consent-adjacent context, it edges toward a genuine ethical problem rather than merely a marketing one.

Anywhere clinical nuance is required.

An AI clone, however well-trained on your material, cannot exercise judgment on a case it wasn't built to evaluate. Patient-specific questions, anything touching suitability or diagnosis, anything where the honest answer is "it depends on your specific case" — this content needs the actual clinician, not a system trained to sound like one.

The disclosure question isn't optional; treating it as optional is the real risk.

Some dentists experimenting with this technology treat disclosure as a creative choice: will viewers even notice?

Does it matter if they don't?

It matters.

Not because regulation will eventually catch up to this (though it likely will), but because the entire value of a personal brand lies in its truth. An AI clone used without disclosure isn't an efficient version of your personal brand. It's a slow-motion way of making that brand worth less every time a patient figures out what they actually watched.

The practices that will come out ahead on this aren't the ones that avoid AI avatars entirely or use them without a second thought. They're the ones being explicit: this explainer was AI-narrated with Dr. [Name]'s guidance; this video is Dr. [Name] speaking live. Patients don't punish practices for using technology. They punish practices for using it dishonestly.

A simple test before you clone yourself.

Before turning any piece of content into an AI-generated version of you, ask one question: is this content valuable because of information, or because of who's saying it?

If the answer is information — care instructions, logistics, general education — an AI version, clearly labelled and clinically reviewed, can responsibly extend your reach. If the answer is who's saying it — a case story, an opinion, anything a patient is watching specifically to get to know you — that content only works if it's actually you.

AI should extend a clinician's presence. It should never quietly replace it.

Everything in this series has made the same underlying point in different ways: patients are choosing a specific person's judgment, not a logo, not a service list, and increasingly not generic content either.

An AI clone is neither automatically the enemy of that trust nor automatically safe for it. It's a tool that amplifies whatever you already are; honest and disclosed, it extends real trust further; hidden and overused, it quietly spends down the exact asset it's pretending to represent.

At TechieDentist, when we build AI-supported content systems for practices, disclosure and clinical review aren't an afterthought bolted on for compliance. They're the actual difference between a system that protects a clinician's trust and one that slowly erodes it without anyone noticing until it's already gone.

Where would you draw your own line, and would you tell your patients when they're watching an AI version of you? Tell me in the comments.