Dentistry

Dentistry is the field that looks after the mouth — the teeth, gums, jaws, and soft tissues that let a person eat, speak, smile, and live without pain. It sits in the Care family because the structural pull at the field level is the same one that animates medicine and nursing: somebody is suffering, and the dentist's job is to end or prevent that suffering. But dentistry carries its own distinct character that separates it from the rest of healthcare. The work is intensely physical, precision-manual, and visual. A dentist operates through a space smaller than a fist, using magnification and mirror-image orientation, with tolerances measured in fractions of a millimetre. The hands matter as much as the diagnosis, and the craft dimension of the work — shaping, sculpting, restoring — gives the field a genuine Aesthetic pull that most healthcare fields do not share.

Resolution is the daily texture for much of clinical dentistry. Teeth break, decay, crack, and fail; the dentist's working day is a stream of problems to be identified, prepared, and restored to function. A filling, a crown, a root canal treatment, an extraction followed by a replacement — each one is a move from broken to functional, delivered in a compressed, hands-on session with the patient in front of you and the result visible immediately. The feedback loop is unusually tight for healthcare: you can see and test what you have done before the patient leaves the chair.

🦊

There's a guide here if you want one

Kitsune can talk through anything on this page — whether it might suit you, what to do next, questions this page doesn't answer. Everything here is yours to read either way.

The diagnostic-consistency floor

Not graded

The barrier

Caries detection has long been notoriously variable between clinicians. The same radiograph read by two dentists routinely produces two different treatment plans, so a patient's diagnosis depended substantially on which dentist they happened to see. This is the honest half of the public's suspicion about dentistry — the variance is real even where the motive is not.

What changed

FDA-cleared computer-vision second reading. Dentists detected 32% more tooth surfaces with caries using Overjet in the study supporting clearance. Five dental AI tools now hold 510(k) clearance for US practices.

Behaviours involved

Finding lesions early enough that prevention is still an option — the Preservation pull the field's centre of gravity is moving toward.
DiagnosingPreventing
A shared visual artefact — the patient sees the annotated radiograph rather than taking the dentist's word for it.
DiagnosingExplaining

What this is based on

Overjet, Pearl, VideaHealth [current_fact as cleared products]
Overjet clearances for bone-level quantification and age-specific caries detection from age four [survey_self_report]

How this could age

Low risk on direction, high risk on framing. That AI reduces diagnostic variance is solid and regulator-examined. Whether that produces better teeth or just more fillings is genuinely unknown. If an independent outcome study lands badly this needs rewriting rather than updating.

What it does not cover

Detecting 32% more caries surfaces is only unambiguously good if those surfaces should be treated, which is a judgement the detector does not make — early enamel lesions can remineralise, and finding them is not the same as needing to drill them. The vendor markets this on a 25% increase in case acceptance, a revenue metric, and secondary reporting notes real-world gains run more moderate than vendor marketing suggests. A tool that improves detection and is sold on treatment conversion has a conflict built into its business model. No independent outcome study establishing that AI-assisted detection improves patient oral health — as opposed to detection counts and acceptance rates — was located.

Assessed July 2026

The dental laboratory's location and craft floor

Not graded

The barrier

Dental technology was a bench craft — wax, plaster, porcelain powder, metal alloys and hand tools. The work required the bench, the materials and years of hand training, tying the technician to the lab and to a local relationship with a dentist.

What changed

CAD/CAM design, intraoral scanning, 3D printing and milling. The design layer has genuinely separated from the fabrication layer; most modern labs are hybrid and the CAD component can increasingly be done remotely.

Behaviours involved

Performed remotely and asynchronously from both the fabrication and the dentist.
CraftingDesigning
Precision and repeatability hand-casting could not reach — implant frameworks and surgical guides previously impractical.
EngineeringMaking

What this is based on

Hybrid lab workflows as the modern norm [inference; Mode 1 narrative]
Custom-printed surgical cutting guides in OMFS, recorded as current practice [inference]

How this could age

Low risk on direction, moderate on magnitude. The digital transition will not reverse. How much of the finishing layer eventually falls to better printing and shade-matching is the open question, and this is the one archetype in the field with real exposure.

What it does not cover

The Mode 1 narrative is clear-eyed and should be respected: the hand skills have not disappeared, and final fit, colour-matching and surface finish still require a trained human eye and hand. This is a collapse of the design and fabrication layer, not the finishing layer. It is also the only genuinely double-edged collapse here — the same narrative warns technicians who do not learn digital tools risk being left behind, which is an opening for one cohort and a displacement for another.

Assessed July 2026

Jobs that did not exist five years ago

These are real jobs that exist now and did not exist before the current wave of AI.

Familiar title, new shape

Digital dental technician / CAD-CAM designer

A substantially re-tooled version of the existing dental technician role, built around CAD design, intraoral scan input, milling and 3D printing, with hand finishing retained.

Mode 1 narrative records the transition in progress and identifies the hybrid technician as current best practice. No posting data located. · early signal

Familiar title, new shape

Digital workflow manager (lab)

Manages the digital case pipeline between clinician, scan, design and fabrication.

Mentioned in the Mode 1 source as a party technicians increasingly work with; not independently evidenced. · early signal

The thinnest new-roles section of this batch, and the thinness is a finding rather than a gap. Dentistry has produced essentially no new roles from AI, for a structural and knowable reason: role boundaries are drawn by the GDC, and scope of practice is a regulatory instrument that does not move because a tool arrived. You cannot invent a new dental role, only be registered in an existing one. The live scope-of-practice fight in this field — hygienist/therapist direct access — is a political struggle about registration categories with nothing to do with AI, and folding it into an AI story would be easy and wrong.

The barriers that recently fell are the other half of this story. They are just above on this page.

What stays yours · where AI reaches

Bars above the line are the parts of this work that still need a person. Bars below it are what AI can already do. Tap any column to see the actual work behind it.

high ground · holds stronglydeep water · reaches furthest

yours, by strengthAI reach, by depth

The honest read. Protection is manual and statutory, which is close to the most durable combination in the set. The AI can see the caries — genuinely better than the dentist, and the FDA agrees. It cannot get to it. The gap between knowing and doing is the width of a drill bit and is not closing, because it is not an information problem. This is why dentistry can have more cleared AI than most of medicine and face almost no displacement risk at the same time: the automatable layer and the valuable layer are cleanly separable here, unlike radiology, where the seeing was the job.

The clearest computer-vision field in the set: a bounded, high-volume, knowable-ground-truth image task. But the field's AI story is not really about diagnostic capability — that is settled and works. It is about what a better detector does when plugged into a payment system that rewards detection.

How AI is changing the way in

2 ways into this field, and AI is not doing the same thing to each of them.

Dental laboratory technicianHarder to enter
The Mode 1 narrative states it directly: technicians who do not learn the digital tools risk being left behind. Design layer exposed (β ~0.5-0.6); fabrication being automated by milling and printing; the protective residue (final fit, colour-matching, surface finish) is a narrower band of the job than it was. Scored moderate not severe because that residue is real, GDC registration gates the title, and the wider shortage cuts across. [inference]
Clinical roles (GDP, Orthodontist, OMFS, Hygienist/Therapist)Largely unchanged
Exposure is concentrated entirely in diagnosis, a step before the work rather than the work. Licensing floor requires a GDC-registered human as clinician of record. Treatment-layer β ~0.02-0.05. Severe shortage. The training pipeline (five-year BDS plus mandatory foundation year; 13-16 years post-school for OMFS) is a far more binding gate than anything AI is doing. [inference / current_fact]

That is everything we currently know about AI in Dentistry. It shows where things are moving so you can choose which way in suits you.

The world−

What just became possible2+