AI In Dentistry L
ast issue we introduced you to a new column we plan to run, Moral Action and Professionalism (MAP) by Dr. Dave Chambers. The focus of
this column is due to Dr. Chambers’ “abiding interest in dental ethics and trying to make contact where it matters most” — thus send- ing out monthly, one-page thoughts about MAP: Practical, actionable, not theoretical. What started with a few friends and added names in response to inquiries and recom- mendations has grown to several hundred email recipients. This issue, we’re featuring two recent columns Dr. Chambers has writ- ten on a hot topic: Artificial Intelligence. On the next page, we include an article featured in the ADA Huddle about the arrival of AI in dental care. So too, the MDA is consider- ing the implications of AI. In fact, this fall, the staff embarked upon AI learning and training through Association AI Professional Certification — a program designed to equip association professionals with both the theo- retical knowledge and practical skills needed to excel in the AI-driven landscape.
LIONS AND TIGERS AND AI, OH MY!
“Toto, I have a feeling we’re not in Kansas anymore.” Every respectable professional needs a little rehearsed patter expressing reasoned concern about AI. The evidence is overwhelming that computers can synthesize diverse sets of existing data. That is not the issue. I am reminded of the debates 50 years ago whether we should allow hand-held calculators in schools.
Yuval Harari, who writes sparkling histories of everything (such as Nexus and Homo Deus), surveyed how civilization and com- munication coevolve. The witch hunts and inquisitions of the fifteenth and sixteenth centuries were struggles about who could say what to others. The printing press was banned in Muslim countries for centuries because authority depended on memorizing religious texts. It is not about the message – it is about shifting power in who controls what others have access to.
28 focus | WINTER 2025 | ISSUE 4
It isn’t the accuracy or the quantity of data that matters. Information is defined as data that reduces uncertainty. And the needs or wants of those whose behavior might be changed by AI is what defined information.
Diagnostic testing and online access by pa- tients to “information” and shared experienc- es with others are the fastest growing uses of AI in dentistry. There is potential in both areas to support changes in the relationships between professionals and patients. At the very least, this is a dynamic market that has the potential to change incomes and status.
Alan Turning, the Brit who broke the Ger- man Enigma Code in WWII, proved that computers cannot do two things: (a) deter- mine whether a computer run is worth doing and (b) explain how the results of the run change any future runs.
He proposed a test to determine whether computers can replace humans (the “Turing Test”). In any case where the user interacts behind a wall with either another human or a computer and the subject cannot tell the difference passes the test. There are two ways to pass the Turing Test: brilliant computers and sloppy humans.
Yes, Dorothy did unmask the Wizard of Oz. It is us.
DENTISTRY IS AI, IT IS NOT OPTIONAL
Arthur Lee Samuel invented machine learn- ing. He taught early (hand cranked?) comput- ers to play checkers by feeding them the best strategies he knew. Most games were a draw, the computer winning when Samuel’s atten- tion wandered. Then he tried something dif- ferent. Instead of telling the computer how to play. He gave it the rules and objectives and let it develop its own strategies and learn from experience.
That is how what we call AI was invented. It is called semi-supervised learning. Give a mechanism: (a) a goal, (b) the capacity to choose actions that lead to approaching the goal and (c) plenty of practice with feedback.
The learning mechanisms in Chinese self- driving cars are every bit as good as Waymo in San Francisco. But the Chinese cars are little threat to the Boston or Dallas markets because they have not been trained in the right context. And Waymo won’t be driving around in Salinas, Kansas any time soon. Dentistry is AI, through and through, and al- ways has been. It is semi-structured learning based on goals, self-correcting responses and lots of experience.
The “machine” in this case is a neurobiologi- cal network that improves the pattern of synaptic connections to optimize fit between goals and outcomes with experience. The dentist’s brain, the office routine and the standards of the profession are AI systems. Dentists have always used emerging support systems to better serve patients. There is not a separate ethics tor AI: there is an ethics for using AI as part of dentistry.
It is all an old idea. Sun Simiao, a master of Chinese medicine, identified the basic formula in the fifth century: “There are fools in the world who read the formula texts for three years and claim that there is no disease under Heaven that is not treatable. And only after they have treated illnesses for three years, do they finally realize what is at stake.”
Sun Simiao was clear that it is the goals that matter. Aristotle insisted that one cannot begin the study or practice of statecraft until one has first mastered the virtues (identi- fied the right goals). Sun Simiao said one
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32
