Very Clinical: Charlie Fryer on the AI in Dentistry Hype Machine

Unknown Source October 14, 2025 36 min

artificial-intelligence generative-ai google

50 Companies

39 Key Quotes

2 Topics

🎯 Summary

Summary of Very Clinical Podcast Episode: AI in Dentistry and Life Anecdotes

This episode of the Very Clinical Podcast, featuring hosts Zach Miner and Kevin Fryer, along with guest Charlie Fryer (Kevin’s son, currently completing a Master’s thesis on AI diagnosis of Class II Caries), blends lighthearted personal stories with a deep dive into the current and future applications of Artificial Intelligence (AI) in dentistry.

1. Main Narrative Arc and Key Discussion Points

The episode begins with a segment dedicated to quirky personal stories about their fathers, establishing a relaxed atmosphere before transitioning into the main technical topic: AI in Dentistry. The discussion centers on how AI is already integrated into dental workflows (like aligner design and digital dentistry software) and explores the potential for broader applications, such as radiographic analysis, implant planning, and practice management optimization. A central theme is the data dependency of effective AI models and the ethical implications of data collection. The conversation concludes with a forward-looking assessment of whether AI will be a net positive or negative force in clinical practice, framed by the analogy of AI as a sophisticated GPS system.

2. Major Topics, Themes, and Subject Areas Covered

Personal Anecdotes: Quirky stories about fathers (farting at the breakfast table, refusal to learn typing).
Skiing Culture: A detailed comparison between Utah and Colorado skiing, focusing on snow quality (drier in Utah) and resort convenience (proximity to Salt Lake City vs. Denver).
AI in Dental Manufacturing/Design: Current uses in clear aligner production and 3D printing workflows.
AI in Diagnostics: The role of AI in reading X-rays and its comparison to traditional algorithmic software.
Data Acquisition and Scale: The necessity of massive datasets (millions of scans) for training robust AI models, contrasting this with the smaller datasets potentially used by diagnostic startups.
Business Implications of Data: The realization that dental companies like Align are fundamentally data/analytics companies, and the parallel between patient data collection and consumer tech (Google).
Future AI Applications: Predictive modeling for patient trajectories (e.g., predicting the need for root canals), schedule optimization, and inventory management forecasting.
Robotics in Dentistry: Brief mention of robotic systems like the Yomi.

3. Technical Concepts, Methodologies, or Frameworks Discussed

Deep Learning/Machine Learning: Mentioned as the underlying technology for pattern recognition in diagnostics.
Recurrent Neural Networks (RNNs): Specifically cited as a potential tool for analyzing patient history trajectories to predict future treatment needs.
Data Points/Scale: The requirement for “millions of data points” (as seen by Align) versus smaller initial datasets for new products.
STL Files: Referenced as the standardized input data collected by 3D printing companies like SprintRay.
Block/Priority Scheduling: A manual practice management technique discussed as a precursor to potential AI optimization.

4. Business Implications and Strategic Insights

Data as the Core Asset: The strategic insight is that companies succeeding in dental tech (like Align) are leveraging proprietary, massive datasets to refine their algorithms, making them difficult for competitors to catch up to.
DSO Data Advantage: The potential for Dental Support Organizations (DSOs) like Heartland or Aspen to become major data sources for AI diagnostic tools, accelerating their adoption and refinement.
Cost Control: The episode features sponsors highlighting the importance of margin management through supply cost reduction (net32) and efficient patient acquisition (Relevance Online Marketing).

5. Key Personalities, Experts, or Thought Leaders Mentioned

Charlie Fryer: Guest, providing insight from his Master’s thesis work on AI diagnosis and his experience in Salt Lake City.
Jacob: Mentioned as one of the few dentists the hosts know who has actually purchased and committed to an AI radiographic tool.
Freddie Clayton: Quoted regarding the predictive power of consumer data (Google) over individuals’ future actions.

6. Predictions, Trends, or Future-Looking Statements

Orthodontics Automation: Zach predicts that orthodontics is “not that far away from being completely AI-driven from top to bottom.”
AI as an Aid, Not Replacement: The consensus leans toward AI serving as a highly accurate “aid” or “GPS” that flags areas for the clinician to review, rather than fully automating decision-making.
Data Exploitation: The hosts agree that users often become “the product” (the data) when signing up for new software platforms.

7. Practical Applications and Real-World Examples

Align/iTero: AI is used to rapidly process Invisalign ClinChecks, likely involving AI generating the initial plan reviewed by a human.
SprintRay: Utilizing AI for rapid design of night guards and crowns directly from intraoral scans.
Radiographic Tools: Software that reads X-rays to flag potential issues (though adoption seems slow).

8. Controversies, Challenges, or Problems Highlighted

Data Source Transparency: A major unresolved question is the initial data source used by radiographic AI companies before they gathered sufficient user data.
Clinician Skepticism: Hosts note that dentists often react defensively or negatively when the topic of AI integration is raised.
Over-reliance: The challenge of using AI recommendations without applying clinical judgment (the GPS analogy).

9. Solutions, Recommendations, or Actionable Advice Provided

Embrace AI as a Tool:

🏢 Companies Mentioned

Heartland ✅ finance/healthcare organization

Mavis Beacon ✅ tech/software (educational)

I GPT ✅ unknown

North Carolina ✅ unknown

Jack Durgan ✅ unknown

Cone Beam ✅ unknown

Maybe Charlie ✅ unknown

Freddie Clayton ✅ unknown

So AI ✅ unknown

So Kevin ✅ unknown

The Farm ✅ unknown

Like I ✅ unknown

East Coast ✅ unknown

Epic Pass ✅ unknown

Alpine Valley ✅ unknown

💬 Key Insights

"So I did a project in one of my classes on predicting click-through rates for advertisements based on like the image itself, like how likely is a user to click on it based"

Impact Score: 10

"It's just going to be an aid, like what you said last time. It's accurate. Just point me of like, hey, you know, you should look at this and that. It's almost like how you approach your GPS when you're driving."

Impact Score: 10

"Align itself is a data company. That's what they are. Sure. An analytic data company that happens to make aligners. That's the fact of that."

Impact Score: 10

"Here's another one we did: marketing, I guess, would be a—I bet marketing would be hugely AI-driven, right?"

Impact Score: 9

"I think one thing is looking at bitewings and doing some diagnoses, but there's a lot of other data that I know you collect with like Cone Beam or the 3Shape, the scans that you do as well. So I think there's probably some next-generation things you could do, just kind of marrying all those different data sets to, I don't know, like the one example, like the guide for implants, like picking an optimal way to do that."

Impact Score: 9

"You didn't know that you were the product when you signed up for Facebook. That's a good point. Yeah, you're right. And so you are just data to them."

Impact Score: 9

📊 Topics

#artificialintelligence 61 #generativeai 3