Subscribe
Share
Share
Embed
The practice ownership podcast for dentists ready to start, buy, or grow their dental business.
Mike Dinsio, MBA and Paula Quinn, BSDH have joined forces, combining their past shows "Startup Unscripted" and "Dental Acquisition Unscripted" into one powerful channel. Together they bring 30+ years of dental industry experience with a modern approach to the business of dentistry.
Each episode features unscripted conversations with expert guests who share real experiences, proven strategies, and actionable insights on dental practice ownership. Whether you're new to ownership, planning a dental startup, or navigating a practice acquisition—we've got you covered.
FRESH CONTENT ON:
✓ How to start a dental practice from scratch
✓ Buying a dental practice: What you need to know
✓ Practice management hacks, tips, and tricks
✓ Dental marketing and SEO strategies that work
✓ Financial planning and profitability for practice owners
✓ Building and leading high-performing dental teams
✓ Industry trends: DSOs, technology, and the future of dentistry
Mike and Paula bring practical wisdom from consulting hundreds of dentists through Next Level Consultants. Their approach? Customized strategies for your unique market and goals—no cookie-cutter packages.
Join our growing community of dental practice owners and entrepreneurs. Watch live on YouTube and ask questions directly, or subscribe on Apple Podcasts and Spotify.
New episodes weekly. No scripts. Just real talk about running a successful dental practice.
Visit: www.nxlevelconsultants.com/resources/podcast
Rate the show 5 stars and help other dentists discover us! ⭐⭐⭐⭐⭐
0:29
Well, Welcome to Dental Unscripted,
0:37
where Mike Dinsio and Paula Quinn break down the practice ownership journey one episode at a time.
0:46
Starting up, buying, and running a successful dental practice.
1:06
All right, here we are. Welcome, welcome back. We are kicking off a week of crazy and
1:16
super intelligent people that will be talking all about technology in the
1:21
dental industry. We have branded this week as shark bite week. B Y T shark
1:29
bite week and we're super excited to be doing it. It's this is episode one to kick off
1:34
the whole week and it's all about AI and dental and what's to come in the future
1:42
and uh I'm super excited. So this is again Michael Dinsio and you your uh
1:47
co-host uh Paula Quinn. What's up Paula? You ready? I thought you forgot forgot my name for
1:54
a minute. No, I didn't forget your name. Are you ready for this week? I am very ready.
1:59
What are you most excited about with with all the folks that we are interviewing? And today is going to be a
2:06
great way to start this program up. Apollo, what are you most excited about this week about? I think I'm most excited about being
2:13
able to bring new things to our clients. you know out there dentists are always leaning on us to what's the latest
2:20
greatest what can we do to improve efficiencies reduce our overhead you know all those different things that we
2:27
get and I think AI is a great solution for a lot of that um use it where you
2:32
can and then obviously you still have the support of your team and and you
2:37
know yourself so it shouldn't change a lot of things it should just make things easier so I'm excited
2:42
Amen I I think that's perfect because like the folks that don't pay attention to this week or or if if none of this
2:51
stuff is on your radar yet, you're behind. you're way behind because
2:56
literally since 2019 20 n 2019 and 2020
3:04
not not 1920 um uh AI has just like totally have just
3:11
totally come in by storm in our little industry of dental and it's already
3:17
disrupting some stuff and I think um I think this week is a is a is appropriate
3:23
because in the next two years it's going to be in our opinions Paul and I next level
3:29
it's going to be like literally the way the way like remember before co people
3:36
were still doing the whole forms in in office
3:41
they were doing the forms longhand now since co and the shutdown it's like it's
3:47
like digital forms are like that's the only way now like like forms remember
3:53
that Paula like it was Like I wouldn't say it's the only way, but it is it kind of is the only way. And if if you can't do digital
4:00
I mean co co literally changed a lot of things the way we operate in the office.
4:06
So I think that we're about to hit this other frontier and today's episode is perfect because we are interviewing a
4:14
friend, a mastermind, a lover of all people, someone that pioneered AI
4:20
himself. Uh he his name everybody's got to have a hobby, Michael. You know, you know, you gotta
4:27
have a hobby. There's no one better to start this week because Tim Rich um has been up to AI
4:34
has been AI up to his uh armpits as he says from the very beginning before it
4:40
was even on any of our radars. So Tim, like welcome to the show, buddy. We're going to talk all AI on this episode,
4:46
but thank you for having longtime listener, first time talker, you know. Yeah.
4:52
Why don't you just like introduce yourself to the audience? Um and um you
4:58
know just give us a little bit about like why we might want to listen to you
5:03
and like why you might be one of the authorities out there like g give us your background man.
5:09
Sure. Sure. Uh well first off thank you very much for having me. It's an honor to be here and I really love uh what you
5:15
guys are doing. Been a big fan of Michael and Paul for years. So everybody out there in TV land, uh, you know, they
5:21
didn't just find me on the street. We've been hanging out. Uh, so my name's Tim Rich. Uh, been in AI for a long time,
5:27
been in tech. Before I got really into this, I spent many years in advertising. And the thing I loved about advertising,
5:34
and I still do, is it is functionally people and it's stories, right? And a
5:40
good ad is really just a story. And it has to resonate with people. And this is actually very very relevant to modern
5:47
dental practices because when you boil it down at the end of the day it's a service industry. You have a service
5:52
which is absolutely necessity. It's a healthcare service so we can just duh clearly necessity but it is really
5:59
wrapped in the consumer and the story. The way you feel when you go into that office and how you're greeted by the
6:04
staff and how you're brought through the processes and right and how you're kept aware and kept a breast of what's going
6:10
on in these machines in your mouth and things. Right? That's all stories. That's all story. It's really the same.
6:15
So, I got into AI about five years ago and I came into it actually um using it
6:22
to help me make data make sense. One of the big problems with computing is that
6:28
it runs data and we don't really always know what that data represents in the
6:33
world. If you think about data, let's call it a menu, right? And you go into a restaurant, you sit down and you read a
6:40
menu. You're reading some data. You're looking at the menu, right? You want to know what to eat? We know when it says salmon what it means in the world,
6:46
right? Oh, beef tartar. Oh, I know what that is. That's old beef. Smoked salmon is beef tartar. Exactly.
6:53
Right. The challenge with data is that when you have ones and zeros or clicks in a data set like an Excel spreadsheet,
7:00
it becomes really hard to know what it represents in the world or what it could do for you or what it
7:06
and what it can do for you. Exactly. So I came into AI to really use that and to learn about how data can help us. And
7:12
I've been in it for many years. And one of the things I really think um so I
7:17
work a lot. I'm based out of New York State, close to New York City. I spend a lot of my time in New York because it is
7:23
the closest to the center of capital and closest to the center of capital is where a lot of innovation happens
7:29
because there's a lot of money to pay for innovation, right? And as we know in the dental practices, there's what you
7:34
do your dayto-day. And anytime you bring in something new, there's a bit of a risk because that innovation, if it's
7:40
not shephered through by the likes of Paula and Michael, who are able to help you navigate that innovation successfully, it could go ary and it
7:48
could set you back, right? And so I always like being close to where these places of new things are happening. So
7:54
out of New York City, I love that, man. I love that. I think I think what the the the first time I uh
8:01
John uh Oh, John. I gotta say, John, we got to get John on here, right?
8:06
Well, John John was the OG. He started this podcast with me like five years ago. So, John connected me.
8:12
Yeah. Yeah. Yeah. He If you go back to the I kicked him off. I kicked off Yeah. John. Yeah. That's called moving up, Paula. That's
8:18
an evolution. That's not going back. Yeah. I'm sitting in John C right now.
8:23
Yeah. We out. So, yeah, he might need to do Yeah, we we opened up uh interviews and and uh
8:31
and and Paula beat him out. That's just how it went. But, um Survival of the Fittest.
8:36
So, John connects me to Tim and says he's working on a little project. And what got my ears ringing was that with
8:44
your acumen and your background, you were able well you you did a a project
8:50
with a a insurance company and they tasked you to help them fight claims to
8:58
get paid. And when I hear something like that, I'm like, I got to know this guy
9:04
because he must know something about how to get claims paid. The opposite. That's
9:10
where my brain goes. This guy helps people or companies fight claims the the
9:16
wrong way. Maybe we can get this guy to help us beat some claims. It was a funny It's a funny thing. So, I
9:23
was working in New York City and it was an insurance company reached out to me and they wanted me to help them figure
9:29
out automated ways to find errors in claims. I mean, errors,
9:34
errors, human errors, just not no macity here, right? Just common errors. just and the interesting part of the puzzle
9:42
was the insurance claim when it is received by the insurance company is a consistent form, right? There are fields
9:48
that have to be filled in. There are dependencies, right? If you uh say you got a filling, did you have a pano done
9:55
before? Right? If you said they had this thing, did they have that? So there's these series of logical steps from an
10:02
insurance company's point of view that need to be reflected on the film and the
10:07
claim to therefore make it a valid claim. And what was also Did you just say film?
10:13
Yeah. Film. Oh, look at that man. Dating me. The digital the images. The images. Pardon me.
10:19
Okay. Thank you. Thank you. Yep. Yep. Yep. Let's keep it current and in case you know in case we have to play this in 2035 on the back reel, you know
10:25
what I mean? Let's keep it current. Um so the interesting part of this puzzle for me though was they had all of this
10:32
data massive amounts of claims both that were successful both that went into
10:37
human look at. So they have three classes of and the insurance companies look at incoming claims and they funnel
10:43
them into three buckets. The first bucket is yep good to go great everything's good you know everything's
10:49
ticky boo out it goes and that's about 80% of the total claims coming in. 80% claims coming. It's good to know it
10:56
doesn't feel like it, but Yeah. Well, of course not. Cuz all we you know what I mean? All we remember is when we stub our toe. We never forget
11:01
that we walk every day, right? You know. Yeah. No, no, no, no, no. It's okay to hate on
11:07
them. It's it's it's it's revenue extraction. Yeah. It's a hustle, but like all Yeah. Well,
11:14
you know, uh so we've got this information and so you have the bucket that comes in that's totally good. ticky
11:20
boo and then you have another bucket that goes to a human to review and then out of that you have a okay this is good
11:27
for one reason or another or this is not good. So in that middle bucket it then goes to right or wrong and then it has
11:33
the objectively wrong bucket. So being able to tell whether or not something was objectively 100% right or
11:40
objectively 100% wrong is pretty easy, right? The 100% wrong stuff is wrong
11:46
number, wrong name, wrong so fat finger a code. Um, just straight up BS like you
11:54
didn't do this right. That's relatively simple. Goodbye. Relatively simple also is 100% right.
11:59
Oh, all the numbers align, all the codes align, everything looks good. It's within the thing. Boom. That it goes.
12:05
The real puzzle is that bit in the middle. And so what we did is we took this massive amount of claims. I think
12:10
we did it over a 100 million claims that went through and had different humans
12:15
looking at them and they were doctors, different doctors looking at them and for various reasons stated or not
12:22
approved or denied. And so we built an algorithm that looked at that middle
12:27
bucket of claims because they didn't want to pay a doctor to look at them. Let's cut the BS, right? They don't want to pay that guy. And we created an
12:34
algorithm to kind of sort these to understand why it worked and why it didn't. And there were three learnings I
12:41
had, Michael. The first one was um holy smokes, this is a really hands-off
12:47
industry for being something that is about our bodies and our overall health and I want
12:52
people to be healthy because then they go get jobs and then they pay their taxes and they pay their social security so when I get old I can have some money,
12:58
right? So I need y'all out there healthy and working, goddamn it. Right. Right. Um so I was really shocked at how
13:05
handsoff it was. The second finding was, "Holy cow, what a radical imbalance in
13:12
knowledge." Um, power in money is often displayed and increased by imbalances of
13:19
knowledge. Right? This is why you can't do insider trading. If your brother who works at the company tells you that
13:25
they're about to sell to Adobe for hundred billion dollars and you go and buy the stock, you get popped for doing
13:31
insider trading because you had an imbalance of knowledge, right? Mhm. That's not cool. It's unfair. It
13:38
distorts the market dynamics. So, I was like, "Wow, there's a real bot imbalance of knowledge here." And the third one
13:44
was um was like, "Holy how would you ever do this on the other
13:50
side of the coin, right? The other that is the dental provider, the your folks whom are on
13:56
your show here listening to this, our clients are our URSTW clients and
14:01
intelligent listeners." Drum roll. So we put together missed it.
14:06
What the drum roll? Oh, it's sound effects. Yeah, I meant I had the button. There we go. Boom. So
14:14
Oh, thank you very much. We started this approach called Go Dental. We are seeking to automate claims and we are
14:21
seeking to try and take the same learning uh through that. Now how do you deal with knowledge imbalance because we
14:28
are we're unable to of course take any work with us nor should we. It's contract based work. They have every right to keep all that IP as they
14:36
should. Fair is fair. Those who pay the piper picked the tune or is that how it goes? And so what we are doing is we are
14:43
working across a number of different offices and different around the United States to help train our AI. So what is
14:50
different now? Okay. Wait, wait, wait, wait, wait. Let's let's pause and take take back.
14:55
So, so that was that was your kind of um and that was when I learned about what
15:01
was going on. Yeah, that was your entrance into dental, right? Like you
15:07
you're like or AI or both or really dental? I I got hired because I had the chops to figure out how to solve
15:14
that puzzle. Okay, that's right. So, you've been you've been the chops the the
15:19
I could my way through the interview enough, Paula, to be able to convince them that I could solve the
15:24
I got it. I got it. I can see that. So you so you had this this background,
15:29
this this algorithm, this AI kind of background, and then this company hires you to do what you just uh complicatedly
15:37
said. Um that's not I wanted to be clear. I wanted to be clear about what we were achieving.
15:43
No, no, no. It's it's perfect. I It's 100% perfect because it is a complicated
15:48
thing. So after you got through that project, you're probably thinking, gosh, like what are some ways AI could uh
15:58
essentially help these these doctors, these practice owners, these small
16:03
business owners. It's not just the guys at New York that need your help. It's it's who really need your help.
16:09
Frankly, they don't need the help. They're doing okay. They're doing okay. You know what I mean? They're doing okay. They're all right. But the but the small businesses do need
16:16
your help. And so, hell yes. Um, so with that that being said, you started dabbling and you started
16:22
thinking about ways to help. Is that right? That is. So uh my brother-in-law had
16:27
been in the is still has been in the dental space for a long time and he started doing uh it like uh PMS systems
16:35
right Eagle helping to set that up helping scale them helping to make sure he's done a bunch of projects around uh
16:41
DDS rescue is one of his projects and right really these very specific solutions inside the dental space and he
16:48
kind of came to me and he said wow it is incredible how many humans are on the
16:54
dental claim creation side. And I said, "Wow, it's incredible how many humans
17:01
are not on the dental claim processing side, right? That is a very interesting
17:09
thing. That's a power imbalance, right? That's a knowledge imbalance. That's a power imbalance, right? So claims claims being submitted,
17:17
processed, and then and then claims being Yeah. processed. Sorry. So submitted and processed. Um
17:23
Yes. versus being created. Yeah. Right. And then as we all know, creation
17:28
is the first step and then validate it. Right? You can make it. That's great. We should check it before we send it in.
17:34
Right? Because otherwise your AR starts to have radically long rolling AR problems, right? 180 days, 300 days of
17:41
outstanding payments from insurance companies. Like dude, there are listeners out there, you got businesses and kids in school and gas and cars and
17:48
people to pay. Like aging AR is a huge problem, right? Because Oh yeah. Oh yeah.
17:54
Right. That's an issue. We've had a few podcasts about that. Exactly. I believe it. I believe it. It's a crazy problem. Right. Because it shakes right
18:02
downhill. So what we started working on as we started putting this together is we started thinking about how could we
18:07
use this new thing called a large language model to help even the playing
18:13
field. Because what I built for the original insurance companies was not a modern AI model. that is an open AI or
18:21
an anthropic a large language model. I ultimately used machine learning and pattern recognition to be able to build
18:28
this but it was a very specific model built for insurance agencies. So it was
18:33
a very specific use case. So you could use very specific mathematics for it. When we get into the real world of where
18:40
AI is at today and what Michael foreshadowed in his uh precient intro is these are called general models, right?
18:47
And what we're talking about here is how do we apply something more general to help this problem of creating claims.
18:54
Yeah. So Tim, go ahead, Paul. Um, did you I thought you were Yeah, I was just going to say so how what
19:01
you're trying to achieve, how is it going to help a dental practice or the front office team?
19:07
The way that what we're trying to achieve is to make sure that what you submit is correct the first time every time. That's it, right? We're not going
19:14
to be able to so that the entry is just as easy as the processing that you need. That's exactly what
19:20
So it always goes in the 80% bucket. It always lands in that 80% bucket. Pay
19:26
it. Every little thing that they're looking to like pinch you on, they change it, right? And we know that insurance companies change their
19:32
qualification for what makes a claim quite regularly. Right. This isn't the same form. Wasn't the form.
19:38
It's kind of cool. You got to see the back end so you know what to do on the front end now. Yeah. Absolutely. Uh it's not so much as
19:45
like stealing secrets or like you know smuggling out the goods.
19:50
It's called experience. It's the same with me. I I worked in a practice. Then
19:55
I bought and sold a practice. Now I'm a consultant that helps dentists buy practices.
20:00
Yeah. You watch that whole arc. I experienced the pain. I try to prevent the pain of them having to go through
20:07
what I went through because of that experience. So I equate it to that. Yes, 100%. Tim, it you know to to you
20:16
you said it best. You said in the beginning of the episode that it's really easy to identify the really right
20:23
and the really wrong. So there's so there's a formula there's a there's a a box there's boxes to be checked, right?
20:31
And so to build a model where every box is checked, how difficult is that? How
20:37
challenging? How easy is it? Is it hard? It's not easy. It's hard. And the reason being is we are in effect chasing the
20:46
dragon. When you think of every box being checked, that means every little
20:51
thing that a human could screw up is known and checked against. What are
20:56
humans really good at? Screwing up. We're aces at screwing up, which means
21:02
we screw up in so many beautiful ways. Each one unique and beautiful in its own
21:07
flower. This comes from the lover of people. That's the love of people, right? Exactly. Exactly.
21:13
Exactly. And so it becomes very challenging to capture each edge case of
21:18
all of the screw-ups because as soon as you capture a set, guess what? Something's outside of it and
21:24
something's outside of it. And that's where we are leaning towards these larger models because they are reasoning
21:30
models. I don't know. Michael, give me a hand signal if I start going too nerdy. But
21:36
what kind of hand signal? Yeah, that one right there. If you do this, if I go nerdy, yeah, I'm gonna be
21:42
like I'm gonna back up off that. But let me just hit it real fast for for for for those of you that are watching on YouTube, he basically told
21:48
us to flip them off. So yeah, that's exactly right. You just send that bird this way and I'm going to switch it up.
21:54
Exactly. That That's perfect, Tim. We'll we'll interrupt you. I I think if if I'm
22:00
following then I think the rest of the viewers are following. But if I start losing you, if you lose if you lose the
22:06
dumbest guy in the room, then you're we got to reel it back. So wait, that doesn't make sense. If you're
22:12
the dumbest guy, I am the dumbest guy. That means everyone listening is smarter than you or dumb.
22:18
Yeah, that's what I'm saying. So everybody else is smarter. I'm judged by you then. Yeah, I'm the dumbest guy. It's perfect.
22:24
How about the dumbest one on the mic? How about that? We'll take that. You're good. You're good, baby. You're good. Okay. Uh, so here's the interesting
22:32
thing of all of this, and maybe this isn't interesting. I find it interesting. Before chat GPT, right? And this came
22:40
out in about 2020. Every computer system was a logical system. Yes. No, and or, yes, part of
22:48
this group, not part of this group. And every computer system that was written was a series of logical commands. That
22:55
was code. The nerds who are writing code in the matrix or in hackers with
23:01
Angelina Jolie, right? They are writing strings of logical statements and the
23:07
computer starts at the top and we're it's like Excel. You can you can say if this then this
23:12
then that boom logical statement. All of computers were created with logical
23:18
statements. Mhm. Large language models that came out to the public with chat GPT in 2020 are not
23:25
based in deterministic thinking. They are based in
23:30
probabilistic thinking. And this means that instead of the Excel formula if
23:36
this then that the large language model is writing if this what is the
23:43
probability of that and go find a hundred examples of that and pick me the
23:51
one with the highest possible probability. It's just like
23:57
it's that is a humongous shift. We That's massive. That's a shift of data
24:03
too. I guess we probably Oh, it's a shift of the way computers are written. It's a shift of the way data is used. It's a shift of the way
24:09
computers use the data. It is the most fundamental change I think that our society has seen since steam. It's not
24:17
the fact that the LLM you could talk to it and ask it how to make a cocktail or that you could ask it, "This is the food
24:22
in my fridge. What do I cook for dinner?" It's the fact that every time you ask it with the exact same stuff, it will give
24:28
you Oh, wait. Can I ask it that? Yeah, I do. I do all the time, dude. Yes,
24:33
dude. I put first I needed to go to the grocery store for me. Yeah. Well, after Instacart arrives and
24:38
you hustle it into the fridge, you know. Okay. So, I I've been doing this one.
24:43
Hey, Chad GPT. We That's not the That's not this episode, but I've been going in
24:49
saying, I have carrots, peppers, onions, this random that, this, that, and
24:54
the other. What should I make today or over the next two days to maximize my refrigerator? And it spits out gourmet
25:02
gourmet recipes and I use up all the like to down to the seasoning. Yes. Yes.
25:08
To this and then it's like you all you need to buy now is this and this and you've got
25:14
three more meals. And I'm like what? It's incredible. So to that point, that's probability.
25:21
A computer nerd didn't sit there and think of everything in your fridge, Michael. And the computer nerd didn't
25:26
create an Excel formula that if tomatoes plus rice then paprika. No, no, no, no,
25:32
no, no, no. Those are all probabilities. And the AI reads a billion recipes and looks at the
25:40
likelihood of rice and paprika and peppers in a recipe. And then you say
25:45
you have margarm or a spice and it goes, oh, here's the most likely recipe.
25:53
What is it? Dude, do we have a sound effect for that? No, we're recording.
25:58
Get one. Next. Next one. We're recording Tim to use that sound next time. Yeah, that's exactly right. Licensed.
26:04
Fully licensed. Next level consultants. Paid in full. Paid in full. So, so just to appreciate
26:11
what you're saying here, right? Like we're about to interview a week of tech
26:17
people. There's some really brilliant people behind all of the services that we are
26:24
interviewing, right? And to just think about those people behind the scenes who have
26:31
created these incredible solutions that we're about to interview. I just want the audience to really appreciate
26:38
the craziness of Tim and this background that they have had to go through. So if
26:44
you are trying something to incorporate in your office and be be one of the
26:49
first movers on these opportunities, have a little grace
26:55
because holy crap to figure this stuff out is great. But but it's happening.
27:00
It's literally happening at a speed that we can't keep up with. So that is the truth. And I think that is
27:07
the most um I think that's where a lot of the agida from society comes from is
27:13
the nature of these models and how we use them every day is just going faster and faster and we are already using AI
27:20
whether we know it or not. My favorite one is the iPhone photo app where I type
27:26
into the search bar dog or I type into the search bar bicycle and it just sorts
27:31
all my photographs and shows me the ones with a dog in it. Right? That's AI. That's computer vision. That's heavy,
27:37
right? That's just like, how good is that? I don't have to scroll and remember when I took that cute picture. I like when you take a picture of
27:42
something and it tells you what it is. Yeah. Google Lens. Yeah.
27:47
Yeah. It's incredible. Right. Like all this stuff is coming in really, really quickly. And I think um
27:55
as much as I would love for society to have the time to understand and
28:03
thoughtfully integrate these things. You cannot put the toothpaste back in
28:08
the tube. What do you mean by that? You're getting real deep on me. I can't.
28:14
Right. You get it, Paula. Like the door is open. You're not getting it back in. Like the the the cows have left the
28:21
barn. The AI is out. People are building with it. The models are open source. You do not need to be a guy who lives in New
28:29
York City and nerds his whole life to build this. You can do it on the cheapest laptop for 400 bucks you pick
28:34
up at Walmart. The best thing about AI is using AI to teach you about AI. I
28:42
learned everything on building AI from literally asking AI. I had it create me
28:49
a sevenmonth educational program. I literally told my chat GPT, I don't know
28:54
anything. I want to start at the beginning. I don't want any mathematics, but I want to understand completely how AIS are built, trained, modeled, and
29:01
scaled into business. Give it to me in 15minute increments. I want you to test me on it every two increments. I'm going
29:07
to take it every day, and I want this to last seven months. Dude, you're a pro. That's crazy. And then it just taught me. And I And
29:14
then it learned how I learned. when I do better, I'm more of a visual learner. It recognized that and it stopped teaching
29:21
me in text and it started making images to teach me ideas. It understood I love
29:27
trout fishing and it started using analogies to trout fishing to anchor the knowledge and knowledge I already had.
29:34
The AI learned how I learned best and optimized itself to optimize my
29:40
learning. I've learned thermodynamics this way. I wanted to know what happens when a jet engine moves faster than the
29:46
speed of sound. But why? Air. Why not? Curiosity is dope. Look, look at him, Paula. He's a mad
29:53
genius. That's I'm just kidding. I'm just kidding. Curiosity. He's like, I don't know what else to
29:58
learn. So, I'm just kidding. Well, everything what I don't know would fill a warehouse. So, there's I got a long list of stuff for me to learn. I
30:05
don't I'm going to die and not get anywhere near it, you know. So So, Tim, let's reel this back in.
30:10
Yeah, let's get it home. Bring it on. Bring us home. How how so so you're you're dabbling with a a little company
30:17
called Go Dental. Yep. Yep. And you're taking Go Dental AI
30:24
Go Dental AI. And you guys are dabbling with this knowledge that you have, the experience
30:30
you have, knowledge, experience, all the things. M and you're trying to put together a a a
30:37
a few services that would change the front office
30:42
forever. And we believe so. Why don't you take why don't you take just a minute uh or a few minutes um
30:50
to describe what you guys are working on and and um and uh where you think you're
30:56
at. Great. Uh what we're working on is we are trying to take the claims but as
31:03
they are put into the PMS system eagleoft dentrics open dental whatever
31:08
right and when it is entered into the PMS system be able to look at it and
31:15
then be able to understand if it is correct or not. And what is different about our approach versus previous
31:22
approaches is that we are looking at it as if we were humans. We have trained
31:27
our AI to literally look at the screen the same way that your front office staff does. It reads the cells on the
31:35
screen the same way the front office staff does. And then it is all about
31:40
putting together those solutions based upon what they see. And this is new because like they just are trying to
31:47
mimic humans. We're not trying to create that checklist that you see that thing
31:52
that you're doing that's written on the sticky note and first you do this and then you do this and then you check this. We are literally trying to copy
32:00
human activity and then it gets even headier, Michael. The AI goes out and it
32:07
runs the mouse as if it was a human, but there's no human there.
32:12
It types on the keyboard. Yes. As if there was no human there. The idea is to
32:18
actually use the front end of the PMS softwares rather than the back end which
32:25
is the databases and the pipes to update this information because the front end of the softwares
32:32
have far more control. M so like you
32:38
know what's interesting is I as I talk about um with the startups before they they even hit Paula's plate Paula Paula
32:45
takes the our clients that are startups in the back end and gets them to open and they always bring up this idea of
32:52
software what software should I go I like this one I watch this podcast I heard this
32:58
person talk my buddies do this blah blah blah blah bajillion softwares out there
33:03
and and And and and we do think I think anybody that's kind of following this stuff knows there's going to be some
33:09
consolidation at some point where Dentric buys them all and it's over again, right? But there there will
33:16
be someone Yeah. Yeah. I'm just saying D there's just too many, right? And and
33:22
the more clouds and all that just it they'll get big and then some will buy them. But the conversation that I have
33:29
with these these docs is I'm like look like AI and softwares and services are
33:35
coming out so rapidly
33:40
in the next whatever years they're out now but there's going to be more stuff coming down the pipe.
33:46
These software developers like you putting together services that you guys
33:51
are trying to put now times that by everybody else that's trying to do it. Who are they coding? Who are they
33:57
creating the software for? And they're of course going after the biggest software companies because that's the
34:04
market share. Yeah. So, do do you want to go down the path of choosing the newest and latest greatest practice
34:11
management software that nobody is coding any of these solutions for, or do you want to go with one of the big uh
34:18
practice management softwares that everybody is coding for right now? because then you're going to be able to
34:24
enjoy those services when they come out. And so like do you want my honest opinion?
34:30
I want I do. I want you to think about what I just said. Okay. So this is not next levels
34:36
certified opinion. This is a Tim Rich opinion. Yeah. This guy that loves people and is
34:41
excited about that. Can you throw that disclaimer up there and make him read it? Yeah. Yes. This is not This is not
34:49
It might be It might be our opinion. We haven't heard it yet. Yeah, exactly. But I just want to say
34:56
Okay, go ahead. Go ahead. Here's the real real. The big softwares
35:01
that are out there are made with deterministic math. That Excel
35:06
spreadsheet if then else. That is dead. It sucks. They're poorly made. They were
35:14
written in 1997. The back ends of them are labyrinthian and shitty and crappy
35:19
and terrible to maintain. The front ends are ugly. Don't tell me you look at dentrics and you say, "Man, that's a
35:26
good looking computer program." No. Whoever made that should be shot and pulled into the ocean. The
35:31
They came They came up with that 20 20 years ago. Yeah. And you know how much money they've made off it and never updated
35:37
it? True. They update it all the time. Yeah. But yeah, they update it. But this
35:45
is my point. Who do you want to put your money with? Do you really want to put your money with a scholaritic beast that clearly
35:51
makes all their margin on glue and goo and not on software? Are you going to put all your eggs into the basket of a
35:58
software company that has no decent software text to make a beautiful front end? Okay. So, who
36:03
is that where we're putting our money? So, who are you backing, Tim? Who are you backing? I don't back any of them. I think the
36:08
future is not a PMS software at all. I think the future what we're actually going to go to is in v a audio interface
36:16
where you speak into the computer and you actually don't put anything into cells. I think the next PMS software
36:23
will not look anything like a PMS software. I think if you're talking like the next thing in the next three or four
36:29
years, I would look for something that has zero look. Is this where we're going in two years? Is this our
36:35
wait or wait? Yeah. Okay. Okay. We're backing off. We're backing off. No, no, no. I love this. Is this our is
36:41
this our where we're going to go in two years? Well, no, that's a different question. This is just about PMS systems.
36:46
Let me say one thing about the PMS systems. It's interesting. I've never put it together like that. But I do
36:54
listen, I I'm I'm not in that world as much as you, you know. I am a little but I'm not a dinosaur, that's for sure. But
37:01
I'm definitely, you know, I'm a hygienist. I worked in dental practices
37:06
cons, you know, we I've definitely had some cutting edge things, but I definitely not out there in the tech
37:12
world. But I do always say, and this is this is no dig on Dentrics, um,
37:18
yeah, no dig on anyone or Eaglesoft, but what I do say is the the negative to those big giants are
37:29
they don't incorporate with other programs. and the and you can't dentrics
37:35
can't be good at everything. For instance, their their patient forms.
37:41
Yep. For for me, dental intel engagement blows them out of the water when it comes to that that front office
37:48
administrative integrating with patients, all that stuff. So, it's funny that they hold on to that
37:57
and hold their audience captive. and you try it and it's not good. It's
38:03
like why not play I I feel like if they play nice in the sandbox more people will stick with dentur eventually they
38:11
are going to outdate themselves because it's they don't have this they don't have that they don't have this or it's
38:16
you know why they don't integrate Paula because the systems themselves are not designed for integration
38:22
to allow a platform like an Eagle Software Dentric to be able to have a different type of dental intake form
38:29
would mean they would have to cut that piece of their software out and allow something to fit in. And it's not
38:35
designed in such a way to do that. But Modento, dental intel will work with Dentric.
38:40
Yeah. Because Modento was made after Dentric. Dentric be able to do it.
38:47
No, but what I'm saying is Drix poo poos a lot of those companies that are
38:52
innovative that come in and try to they'll only let them do so much. Do you know what she what she's saying is is like
38:59
look like if they I don't think it's a business decision. I think it's a technological reality
39:05
that is being sold as a business decision. What you're saying Paula is, oh, this is a strategic play on the part of Dentric.
39:12
No, I don't think their technology would even support it if they wanted to.
39:18
Oh, I see. And if they wanted to, it's so crappy and bad, they'd have to nuke the whole thing, spend 5 years and 50
39:24
million bucks to rebuild it to be able to enable that business, which which they've made, by the way.
39:30
They've made it. They've made it. They just didn't do it along the way. Yeah. Yeah. It reminds me of
39:35
Why would you want to refactor that? Why would you shoot the goose that made the golden egg? Keep it laying golden eggs until it ultimately dies of old age and
39:42
all your shareholders go away and you make your 50 mil, call it a day, right? There's no looking forward. It reminds
39:48
me of the city planners of Seattle. That's what it reminds me. Yeah. Running I5 through the city of
39:53
Seattle. That was really great in 1991. Yeah. When the property value was at $20,000 an acre. That's really great.
40:00
Now the property value is at $20 million an acre and you got to put a new lane and it's not going to happen. You can't just build around like most
40:06
No, they have to go down in the ground or up because there's too much water.
40:13
Too much water. It's all water and hills. Seattle actually used to be 200
40:18
feet higher, right? Capitol Hill was cut off and dumped into the bay. Downtown was actually the top of Capitol Hill
40:23
that they cut off. They remove vertical feet. I do know they have like a whole city under the city that used to used to be
40:31
the that they backed in. They filled it with Capitol Hill. Really? Yeah.
40:37
All right. So, let's move back to AI and get off of Seattle. Sorry. Seattle. I was the I was the one that did that.
40:43
I got my masters in Seattle. I got one of my M. Yeah, I spent some time there. Go. You go. Go Dogs. So, so how would you put like I knew
40:52
this was going to be a hard interview because Tim's brain is so much bigger than most of ours. And so I want
40:59
Hey, now. Well, maybe not. Paul, I put on my pants one leg at a time like everybody else. Michael,
41:05
I wanna I want to try to bring this back. See, he's already trying to he's already trying to um I I want to
41:12
Is he gonna take my job or you just gonna like I took I took John's job. Is
41:18
See, Paul is half the problem. See, see, she's taking it down another another path again. Okay, I'm trying to put this
41:25
in bite-sized pieces and control the cat. For any poor dentist that's trying to listen to this,
41:30
that's trying to listen to this whole madness. I how can Tim I'm going to ask
41:36
you a question and and we've as we're going to ask every single
41:41
interview for this week. We're going to ask one question. Let's let's put a bow on this episode by by asking you this
41:48
question. Someone that's in Den now working on software. You're literally working on software. We're going to
41:53
interview some other folks that have their software and and it's working great. Um, we've got uh X-ray companies that
42:02
are reading X-rays with AI. We got AI agents listening and answering calls. We've got uh chart perio chart AI um
42:11
amazing um company that's that's doing all that. Like we've got just some really cool stuff coming down the pipe.
42:18
My question for you, sir, is in the next two years, and I want you
42:23
to think about that. Two years, not He's not going to think about it. Not 10 years, not three years, not well maybe
42:29
one year. Well, let's just say in the next 2028. In 2028,
42:35
how is AI going to change our dental
42:40
ind? How is our dental industry going to be changed by AI in the next two years? So, it's it's
42:46
probably going to be even crazier. We talked about some stuff in 10 years. Apparently in Tim's world, you're just
42:53
not even gonna have a PMS and you're just gonna speak to a computer and it's going to do everything for you in 10
42:58
years. But in two, you can quote me on that. The next PMS won't be a PMS. I'm gonna quote
43:03
that's really that'sing looking at the bill. In the next two years, what would you
43:09
predict? Go. How in the next two years, I think we are going to see two distinct changes happen
43:16
within the dental industry. I love that. And the first change that we are going to see is we are going to
43:21
see a way that data the patient data is
43:26
moved. I believe that the calling up and sending the records by fax or by email
43:33
or whatever. I think we are honestly going to see a centralized patient data
43:38
exchange. Wait to referring uh specialists to insurance companies.
43:45
Remember, one of the P's of HIPPA is portability. And today, there is no
43:50
portability of data. That's true. Okay. One of the It's like whoever whoever does port it
43:56
is in trouble cuz someone's going to put a fine on you. That's right. Exactly. Right. But it's literally So, I
44:02
think one of the things we're going to see is radical portability. We're already seeing in medical, right? I was just going to ask, are they doing
44:08
that in medical? No, they are doing that much easier in medical, right? And Michael, one of the
44:13
forces that's going to make this happen is the centralization of the technology. Your initial insight of how companies
44:18
are going to glom together and get eaten. Ultimately, it will get eaten and it will be a larger company that rams
44:24
through portability and it's going to be a Google or it's going to be an Amazon. Massive behemoths.
44:29
That's amazing. You you lost me a little bit with the forces of nature and I thought of Star Wars and the good and
44:35
the evil. and his his brain started going like lightsabers and he's like what color lightsaber would I have
44:43
okay number two the second thing that I think we are going to see is I think we are going to
44:49
see far more automated analysis of what's going on in the mouth for example
44:54
went to the dentist today as a matter of fact in prep for this interview I wanted to go I wanted to be close to the source
45:00
and they went around and they looked at the bone density and they did all the numbers 3 2 3 1 3 2 3 1 and I thought to
45:07
myself, That is totally subjective by the nature
45:12
of this dentist. She's Oh, wait. There is one. There is one. You got to We are going to see that far more. And
45:19
that is just one example. We're going to see. Wait, you're saying subjective? The the person still doing it, you're saying
45:26
that gets reflected. Yes. 3 2 1 3 4 3 1 and she's literally
45:32
looking and doing the bone density thing. Okay. What do you think? You think they're just going to scan it
45:38
to scan it? Boom. They're going to do a density. They're going to do some kind of wave that they're going to cut through it. But what I saw was not that
45:44
it was bad, but what I really saw was somebody who had had her coffee or not. Someone who liked me or not. This is
45:51
really, this is the cop gives you the ticket because they like you or not. We all do. 65 in the
45:57
You know what's funny is I'm not gonna say who. I'm not going to say I'm just going to be real quiet. But I
46:05
200 I would say 16 pretty sure 2016 I can
46:11
find my email. I introduced to a company the idea of this of scanning equals
46:21
parocharting. Yeah. Perio charting. That's the word. Thank you. I have my email.
46:26
So do I get any credit? Do I get any credit when somebody Dental Unscripted says that Paula Quinn
46:33
invented digital definitely gets credit. She definitely gets credit and residuals.
46:39
I have an email that's stamped. That counts. That counts.
46:46
Yeah. Did you So Tim, let's bring it back. Sorry, you guys. So So what you're
46:51
saying is is it may it and it might not even be the perio charting. Let's just back up a second.
46:57
It will be the observation in the world which is the density of bone at the point where the gum meets the tooth
47:03
right that's what was being quantified you call it whatever the heck you want but I think we are going to be automating these types of things the
47:10
bite wings all the panels so where the technology that we have in our practice
47:15
the actual machines the equipment meeting
47:20
the interaction into the software or the claims or whatever like that fully integ
47:27
as a human. What you're missing is the human, the teeth and the mouth. I'm not talking about software and I'm
47:33
not talking about PMS's. What I'm talking about is the fact that the data comes out of my body
47:38
and it's read into the machine and it's used as part of the visuals through scans. It's with her and the perio doing
47:45
those great counts. I love my dentist madly. It's them looking at my crowns and my fillings and going, "This is an amalgam and that's a blah blah blah. Oh,
47:51
look. I see an olusion." The point is is you're capturing data on a human and
47:57
then that goes into this whole digital thing and it's a PMS and it's a that and it's an insurance claim blah blah blah. But the point of collection
48:04
I think is what will be automated the most to put it in dummy denio. Uh I got it
48:10
too to put it in dummy denzio way if medical is doing that. You're right
48:15
because there there's they do a body scan. I just do the I do that $2,000 body scan. I do it every six months.
48:21
Tell you your my statistical likelihood probabilities
48:26
again, right? Not determinism, probabilities coming back. But to take hold on,
48:32
Michael doesn't like us. No, you got you guys are You want to do a dental podcast with me?
48:37
You guys are literally We can cut this Michael guy out, you know. So, I go to a general checkup medical
48:46
and it's I think it's multicare. If you're in multicare the system and you go to a specialist or anything else
48:53
within multiare the information is passed and is synced so much easier
48:59
unless you go to a different network right and and it is super strainless the
49:05
billing the information the records I I had a surgery they the last surgery was
49:12
in there the doctor looked at it I felt like I was being taken care of like amazing because there was that
49:18
integration. I think that's awesome. I think that's really cool for you to that's really fascinating stuff. Imagine
49:24
a dentist and a a a specialist, a GP and a specialist all being in one cohesive
49:32
place with all that data. And then technology even puts another spin on. There was a study that came out six or
49:37
eight months ago that said that there's a higher statistical likelihood of identifying colon cancer through your
49:43
mouth than through a colonoscopy that there are signs of this
49:49
just so you know that's that's most diseases. I agree. I agree. I have always thought
49:55
and I know we're wrapping up because that's your last one and Paula gave me the wrap-up sign but I No, that was to Michael. That was
50:01
Oh, that was the Michael wrap-up sign. Yeah, that was this to Michael. But oh, thank God because I'm about ready to fire up. Um,
50:09
you know, I really have always wondered why is dentistry separate from medicine?
50:15
Well, so yeah. So that well like you know is this I thought it was
50:21
like 1820 18. It's only in the US though. It's only in the US really and truly that's why we
50:27
came out with the APA new classifications because now we're grading and staging. In the beginning
50:33
was it split? I don't you know I always hypothesis Paul whenever I talk to hygienists now I'm
50:39
like holy be the body and the mouth are connected. It's crazy. Did we just
50:45
discover that? Why did we split it? But I have a hypothesis. I don't know. Here it is.
50:50
Why? Why didn't we Why was it split originally? And I think what it was is when the American Medical
50:57
Association was founded in the late 17 early 1800s, they came out of Europe,
51:03
dentistry was having a shot of whiskey, biting a piece of leather, and then they pulled your tooth out with a pair of
51:09
pliers and cleaned your shoes and the doctors were like, "No way." They gave you a haircut and gave you a haircut. And they're
51:14
like, "No way in hell those hacks are hanging out with us. We went to the sorebone and learned it." And they split
51:20
it then. And then ever since then we have been on this course of divergence and we are finally starting to come back
51:27
around. This is a whole another story. We have taught patients to treat us that way
51:33
too. 100%. We have perpetuated the artificial split that happened back in the day
51:39
for sure. I that's a really interesting hypothesis. Probably spot on. You said
51:47
it's coming back in. That we'll save that for another episode, but like yeah,
51:52
that makes me nervous. Be as as someone that doesn't like consolidation of
51:58
dentistry. That makes me nervous. We're not talking about consolidation of dentistry. We're talking about holistic
52:04
medicine. No, we're talking about But you guys But but you guys are being naive that business and and health
52:12
aren't aren't the crush the same. But but hey, if I don't have to have a colonoscopy
52:17
and they can look at my mouth and figure it out, count me in. Yeah. You don't have to drink the stuff and blow up. You don't have to drink the stuff. Sit over
52:23
there like that. Now say they're now realizing that the mouth is the gateway to the body. It's
52:30
connected to the body. You cannot have a healthy body without a healthy mouth. Vice versa. I It's earthshattering the
52:37
things that you can find in somebody's mouth. the bacteria, the the soores,
52:43
different things that tell you the diseases they have in their body. They just never look in the mouth,
52:49
the nose. And they I guarantee medical doctors besides besides going like this, ah, they don't
52:55
they don't look in the mouth. No. And and we don't pay attention to the body. 100%.
53:01
It goes both ways. a data will bring it back together because if you get enough data and you throw enough algorithms on
53:06
top of it Paula don't don't necessarily don't encourage him Paula don't
53:12
encourage him he's going to keep going hold on if it goes back together we're
53:17
talking about doctors who don't get paid enough because they are in hospitals and
53:23
hospitals we all know don't take care as good as private practice facts
53:29
I'm not trying to go down a level of care. I'm just trying to look at what's going on and understand how the forces
53:36
may play out. And that was the question. That was the question. You'll be long gone, Michael. Don't worry about it. Yeah. No way. He's going to get that
53:43
super care. He's got Seattle good love. He'll be all right. Yeah, exactly. He'll never be again.
53:48
Yeah. All right, guys. We got to end this at some point. I know. We could keep going. Yeah. The
53:55
next one. Hand me for the next one. People probably already jumped off. They're like, "I can't. I can't."
54:01
It's very true, but I I pull somebody's teeth out with a pair of pliers and give this guy a haircut. I got to get out of here.
54:07
This is This is exactly how I thought this episode was going to go, guys. We're We're kicking off Shark Bite
54:13
Week. Follow every episode. Watch everyone. Get on board with technology or you will be a dinosaur in in two week
54:20
two weeks, sorry, two years. I fully believe this. Tim's predicting the
54:26
entire Star Wars uh integration of both forces. Um and if you don't jump in,
54:33
then you're going to be on the dark side quick. So, with all that being said,
54:38
Tim, thank you so much for your time. Paula, Paula, Michael, so thank you so much for having me. It's an honor to be here and
54:44
just to be able to talk about this stuff is great. Thank you very much. We really enjoyed it, Tim. Take care. Bye.
54:59
Thanks for listening. Let us know how you like the show. Rate us on Apple and Spotify.
55:08
Subscribe and follow for more.