Artificial intelligence could change dentistry in real ways. It can analyze radiographs, flag patterns of decay or bone loss, assist with treatment planning, help produce patient education. But there is something at the center of the work it cannot replace, which is clinical judgment that incorporates context. I was reading about AI applications when I thought of a particular patient. She had severe wear on her lower front teeth in an unusual pattern. A pattern-recognition system could identify that the wear was present. But finding the cause required conversation. It required understanding her life. I asked about her habits. She ground her teeth at night, had for years, and the grinding had produced the wear. Treatment had to address both the grinding and the damage. But understanding the grinding meant understanding that she was under heavy stress, working two jobs, supporting family, anxious about her health. The wear was the visible end of a long chain that began with her circumstances. No system analyzing the image would have understood that her stress was the root of it. This is what clinical judgment actually is. Integrating pattern recognition with contextual understanding. Yes, there is a cavity here, but why does this patient have cavities? Poor hygiene? Diet? Access to fluoridated water? Systemic disease? Medication that dries the mouth? A system can identify the cavity. Choosing the intervention requires knowing the person and what they can realistically manage. My training in Russia drew a hard line between diagnosis and treatment planning. Diagnosis identifies what is present. Treatment planning decides what to do about it given the specific patient in front of you. A system could become superb at diagnosis. Treatment planning is a different thing, requiring judgment about what is feasible, what the patient values, what matters most among competing options. So I think AI belongs in dentistry as a tool that supports clinicians rather than replaces them. It could flag likely problems on a film so nothing is missed. It could surface differential diagnoses a clinician had not considered. It could help calculate trajectories in planning. But the final decision should stay with the person who understands the patient's full context. The distinction the AI literature draws between narrow systems that excel at a defined task and general intelligence that integrates context and judgment is the whole point. Narrow systems are within reach. The contextual judgment dentistry actually runs on is not. Dentistry needs that contextual judgment, not only pattern recognition. It needs understanding of the person and their life, not just the patterns in their mouth. The future worth building is not dentistry without dentists. It is dentistry with dentists who use these tools skillfully and still make the judgments themselves, having taken the trouble to understand the person attached to the teeth. That judgment is the part no one has automated, and it is the part worth protecting, which means using the tools without letting the skill of thinking for yourself quietly waste away.
