Will AI Become Your Primary Care Doctor? with Hunter Ziesing, BSBA

Will AI Become Your Primary Care Doctor? with Hunter Ziesing, BSBA

00:00

Hey everybody. Welcome back to the energy MD podcast where we help you resolve your long COVID and chronic fatigue syndrome naturally so that you can get back to living your best life. So really excited today cause it kind of marries my interest in AI with healthcare. And we're going to be talking with Hunter Ziesing, BSBA today. He is the co-founder of longevity health and AI powered online medical clinic.

00:28

democratizing personalized preventive healthcare. After spending nearly 17 years in investment banking at bank of America securities and founding conquest capital hedge fund hunter pivoted to healthcare innovation following personal losses to preventable diseases. He founded the nonprofit charity of choice and served as head of partnerships at pace line, learning how to motivate lasting health behavior change through gamification and wearable technology.

00:57

Hunter co-founded Longevity Health with the head of UCSF San Francisco Emergency Medicine to make billionaire level health optimization accessible to everyone through comprehensive testing and AI integration. Hunter, thanks so much for joining me today. Thanks for the write up. Yeah. Excited to have you with me. So let's kind of get in. Let's, mean, let's start very basic. Can you define AI for us?

01:26

uh Define AI. You'd ask me to describe a large language model. Yeah, I mean, however you see it, what would be like accessible to our audience so they can kind of because everybody's got a lot of different thoughts on what AI is and how it's affecting us and you know, does it belong in health care, which we can talk about in a little bit. But how would you how would you define it in simplest terms? So in broadly speaking, it's a tool. For information that.

01:54

is somewhat omniscient of any piece of data that's publicly available. um So I would call it a very sophisticated search engine. Now, that's the application that we're using it for, and many of your people on this call use it for. um But in terms of other use, you're reading about all the thousands or tens of thousands of verticals where they're using AI to uh make

02:24

uh Business more optimized, right? And so we're also going to be doing too Right. So a lot of people are using AI right now also for their health conditions, you know, they'll say hey I'm having these symptoms. What should I do? Right. Yeah in fact it's funny people are becoming their own doctors and That's why I'm so excited about what we're building for consumers is that it used to be uh

02:54

you know, 30, 40 years ago, you get sick, you go to the doctor. 20 years ago, you get sick, you start searching, you know, Google or Yahoo, whatever you use at the time for your condition or what you might have. And it's kind of scary because everything you read says you're going to, you know, you start reading all the symptoms and you match all the symptoms and you're like, I'm going to die. Right. Or I've got a major problem, but it's gotten a lot better. In fact, I have two take home messages for people using

03:25

you know, AI for healthcare or predictive stuff or just figuring out what's going on in your body. It's great. It's 90 % accurate and people are becoming their own doctors. you know, I've had a couple of things come up in my life in the last year and I've used it it's been remarkable. It's made some pretty interesting recommendations. Of course, my doctor refutes a lot of it because they're not paid for a lot of the modalities that it suggests, but we're finding AI is getting more and more accurate.

03:54

In fact, there was a study put out by Stanford and Harvard about a month ago for diagnostics. uh doctors and physicians alone are accurate about 76 % of the time making diagnoses, and uh AI is 92%. Wow. Yeah, and I think that's- sorry. And the take-home message is you don't need to spend a lot of money.

04:21

on all these functional testing companies when you have issues, you can use Chatchie EPT to get most of the way there before you go back to your doctor. Yeah, and I would agree with that. I my whole thing, I do a lot less testing than I used to do. I would say 90 % of all the different causes that we're looking at and people who've got chronic fatigue and long COVID can be determined by their symptoms, by their health history, does not require a lot of labs. And so it makes sense to me that AI would be good at that.

04:50

That's the definition. And what I see as part of the challenge is kind of like treatment that's beyond conventional medicine, because it seems like AI still kind of likes to stay in conventional medicine. If you ask it to be more of a naturopath or a functional medicine provider, it does a pretty good job of it. But I find that it never goes into kind of like the environmental medicine or like the further depths that even if you kind of like ask it to, because it doesn't, when I, I've been testing it to kind of see like,

05:20

can this help with somebody who's coming into my program and replace me, to be frank, so that I can help to kind of create an AI that can potentially replace me. And I'm just not getting to where it needs to be yet, and maybe I need to do some additional It's not ready yet. AI's not ready yet. m Consumers aren't fully ready for the driverless doctor yet. Maybe in San Francisco, you'll get it in a Waymo. But in Seattle, if you saw a driverless car, you'd run for the curb. So consumers aren't quite ready for that.

05:49

um to the internet on the internet. AI is not fully ready yet because it needs to be trained. And that's something we're building now is we're trying to train these models, Gemini, Chachi, PT and others, with a rag on top of it to effectively make it more accurate, acquire more data relative to your specific situation. And then the third um and make accurate recommendations and eventually prescriptions, right? um

06:17

And then the third reason it's not fully ready from a regulatory standpoint, it's just not quite, even though it's super accurate, it's just, know, the, feds are not really ready to, know, put you out of business yet. So I think for the foreseeable future, there's going to be a doctor or always going to be a doctor in the loop, but AI can get a lot better if we train it. Right. And so that's what you're doing right now. You're integrating AI with

06:42

clinical biological data, right, to personalize somebody's experience. Can you kind of tell us about what you're doing? Sure. So oh we have three steps to our process. The first is our major clinic. We have an online longevity clinic. Online being, we don't have no bricks and mortar, no facility, but we have HIPAA compliant doctors. We're in five states right now. We do a very complex interview with the physicians and the clinicians.

07:09

and so what your objectives are, what you're going through, whether it's chronic fatigue or any blood disorders, anything you have going on in your life and what your objectives are. Then we do a series of tests. change, or right now our basic tests are Dextre scan for uh bone density and visceral fat. It's much cheaper than an MRI. We do comprehensive blood work. We give everybody a glucose monitor. We do a sleep study. We actually do a VO2 max test, which

07:36

is not just for the elite athletes. A high VO2 max corresponds with high longevity. That's five. We do a sleep study. There's one more, escaping the top of my head, will come to me in a second. We look, and then in some cases we'll do, when people wanna do additional testing that's specific to them, we'll do it. That's out of pocket. uh We then retest each quarter. We develop a plan for the patients and then they get a Slack channel, however, which they want to communicate.

08:05

and they uh get a dietician, a clinician, an exercise physiologist, and the doctor who can prescribe medicine. um Then we retest every quarter, not all the tests, but we retest every quarter to see if there's improvement. So the whole idea is to help people uh change their behavior and show improvement versus spending 20 or $30,000 on a bricks and mortar clinic where it's kind of a one and done, you go in and they wine and dine you. So that's phase one.

08:35

Phase two is how can we take that experience that you would spend $100,000 on with Peter Atiyah that we charge $10,000 down to $1,000 a month. And we're testing that right now with a small paid beta trial. And like our billionaire bundle, the $10,000 product, we store all of your health data on an AWS secure server, HIPAA-compliant server.

09:04

and uh you initiate the conversation with an AI agent. In fact, we can call it in a second and I can show your users how it works. analyzes all your data, makes recommendations, does exercise programs, uh follows you. It's kind like your personal AI. It's not a coach, it's your AI doctor. um That includes a physician, a doctor in the loop who's checking all that data. So it's not just turning it over to AI. The third...

09:32

you phase of that, and I don't know if it's a year or two or three from now, will be AI only. So your primary care provider will be replaced by AI. Fascinating. And you said $1,000 a month, but I think you've been $1,000 a year, right? $1,000 a year, correct. Thank you. Yeah, that's really fascinating. And so you said it's more of an AI doctor than an AI coach, but it sounds like it's, is it coaching you on the daily? Like you wake up and it's telling you,

10:00

Hey, don't forget to eat this way. Don't forget to take your supplements or your pills or... That's a great question. And that's something that I really want to build into this because my be all end all for this platform and my dream is a billion users on this platform for pennies a day and have it track you every day with an app that has a check engine light on it. So when you're behaving poorly, it'll let you know and you can adjust the, you know, how it reacts with you. It can be really hard on you and call you a fat.

10:30

You know, piece of crap, you know, it knows when you're traveling, hopefully it'll be following your diet, but it'll also give you, you know, subtle fun, friendly reminders. And then, then, you know, the funny thing is the, the, the people in this country that have all these problems, health problems generally want to do something about it. They don't have the tools, they can't afford it and they don't really have the motivation. So to change.

10:57

You know, doing this, these testing, or if you have say chronic fatigue, you know, you have a problem, you, want to do something about it, but a lot of people don't. So getting back to your idea of daily metrics and following you absolutely. And I also think there's an opportunity to gamify it. So let's say, uh, women's over 45 with osteoporosis, um, of, you know, Bellevue, Washington, right?

11:22

Maybe there's 30 people there. They all get together, they're exercising, they're working out together, they're watching each other. It's sort of like Strava challenge uh meets healthcare. There's also ways to provide people financial incentives and rewards. uh I mean, I go on and on and on, but I've got a great idea and I didn't think of it, but doing partnerships with big uh facilities like Lifetime Fitness. And the more you go and the more you work out, we'll track all your data, the less you pay.

11:52

Wow. Yeah. That's pretty cool. know there's some health insurance companies that have done something like that. Like you get discounts the more that you're working out or exercising, but I don't think they're, I don't think they're. I don't think they do a horrible job. It's got to, it's like, you know, when you walk into a hotel room and there's five remotes on the, on the credenza in front of the TV, you don't even turn the TV on or Airbnb, right? Right.

12:16

But now, remotes are smart. They do everything for you. It's got to do everything for you. It's got to be intuitive, easy and fun and gamified to use. Absolutely. Yeah. So how do you, how do you gamify something like that? I mean, you talked a little bit about it, like in terms of incentives, um, anything else, um, like in terms of like motivation, um, that can really help people get off the couch. Yeah, we had, um, a pace line, my former company that I was with for the last two years, it was an app that

12:44

you wore a Garmin Fitbit or an Apple Watch and it tracked your heart rate. And each week you elevated your heart rate to 150 minutes of cumulative heart rate or exercise, which can just be walking the dog at a brisk pace. You would get a reward of a dollar, like an Amazon gift card. And then we had this whole partner platform, like 30 % off Lululemon, right? If you get to a certain streak. And if you missed your streak each week of 150 minutes,

13:13

it would go back to zero. And people would like, on Sunday afternoon when they were at 140 minutes and it was about to get to 150, they would run outside and exercise. So we really did change behavior with that reward system. And there was some social aspects to it where you could compete against your friends. The challenge with that was that most of the people that used the app were already healthy. ah

13:37

So that's one, two, you know, employee HSA plans, we're getting ready to do a large pilot in March with an employee employer. And we're gonna try to layer in some incentives for people to track them. And not just like, hey, here's a gym pass, uh go work out, right? Here's what's going on in your health, and here's some incentive to go do something about it. Because ultimately, this will bring healthcare costs in society down substantially. Yeah, yeah.

14:07

Wow, that's really great. And so then in terms of like creating this app or the next step, which you're talking about, like doing the health coaching, um know, are you, obviously you're doing the programming or you've hired somebody to do the programming that you're layering on top, like you said, a rag on top of the LLM? Correct, correct. And we're in the process right now of doing a small capital raise with investors to...

14:36

bolster our engineering team and start really digging into uh training the LLM and making the whole user experience better. But we were in no rush because we're still waiting for a few things to catch up with AI. And we also want to validate that it's something that consumers really want, which we did. And so how do you with how fast AI is rapidly evolving and changing and how each one of these different LLMs

15:04

ah Is evolving. How do you choose one? Because you have you probably have to choose one and stick with it Like if you're gonna go with chat GPT, you probably have to stick with it now for the long term because you're gonna build everything on top of it, right? Yeah, and I'm probably getting over my skis and answering the question right now. We're working with Gemini and um Chat GPT but I'm gonna stop there because uh my knowledge of that is beyond my

15:31

capabilities from an engineer standpoint, but we've got some exceptional people working on that right now. Nice. Yeah, it's been really interesting to kind of like watch the race and to switch back and forth, you know, as so-and-so comes on the line, Gemini has all these new features, I try them out and then compare them and see what kind of results I get. So it's really interesting to kind of think about it in the context of, okay, how do you choose one of these? then...

15:57

How do you put the information, how do you train then what you want on top of it, knowing that it's just gonna change in another week? Yeah, and who knows where the leaders of these companies are going. I guess Sam Altman's been saying that, know, uh they're gonna replace a lot of the tools that we use, you know, like shopping. um But I don't think any of these companies, including Google, Microsoft, Apple, are going to

16:27

build something that's clinical, nor do I think they'll do it with AI. I still think you need this human element on top of it. I could be wrong. And I don't know whether or not this is your expertise or you feel comfortable diving into this, but I'm really curious about when people say that AI is gonna take jobs and then people are really worried about losing their jobs, do you have any...

16:52

message of hope for them, or what do you think it's really gonna look like in the future? Are people gonna lose a bunch of their jobs, or are they just gonna pivot to different jobs? I think maybe there'll be a bump down in employment over time, but I think it'll pivot. I I remember when Uber came around, all the drivers revolted and tried to stop it in all the cities, and now every taxi driver and more, it's created more demand, right? So there's even more drivers now. um It's probably not the best.

17:22

Example, but you know the buggy whip example, right? I don't know that one. Oh, ah you know when when they invented the car The buggy whip went out of business because nobody was using a horse and carriage So people just pivot and but I do I do think in fact, it's all right It's already you know companies have been reducing headcount You know through this this bubble for the last three years and I think it's gonna get worse before it gets better Yeah Yeah

17:49

Yeah, it'll be really interesting to see. I just have to remain hopeful and believe that, you know, everything's always working out and that em our lives are going to be better with AI as kind of like an adjunct and a support and assistant. I know for me in my business and my personal life, you know, I was just over in Europe traveling and I miss a train and, you know, Google is great as a search engine.

18:16

But now I have the ability to really get a better sense of like, what am I supposed to do? I'm in Italy, I miss a train, I miss our connection. Like, what are my options? And so it's just uh allowing us to get, think, some of this data really a lot more quickly and allowing us to make better decisions for our, at least for our personal lives, and then also our businesses. And I can just hope that it's going to support everybody and everybody's business in this way and that everybody's taken care of.

18:46

So I haven't read your book, I see it on your shelf, I'm gonna have to get a copy of that. And I don't know much about chronic fatigue. um Although as I tell everybody, the three things that have helped me through life and many of my friends that I've helped is exercise, diet and nutrition. So I'd be curious as to how much exercise comes into helping people with chronic fatigue. I mean, and when you're fatigued, you don't wanna exercise.

19:13

Exactly. Yeah. So I tell people that they should get their Goldilocks dose of exercise. It's just like the, the amount that, doesn't make you feel worse because if you exercise and you feel worse, um, that's just, it's too stressful on the body. you know, exercise has got this great hormetic reaction where it stimulates us in a way that's, that's challenging that allows us to build back bigger, you know, like a muscle. Um, but unfortunately, hormesis is also a stressor.

19:41

And it's too hard on the people that we work with in many cases. so it's finding that Goldilocks dose. But there's still so many other markers that people can be supported with in terms of moving forward. And exercise is still important, or I call it movement, because sometimes it's just about dancing for five minutes a day. Yeah, interesting. And you told me earlier, most of your uh clients are women. Yes.

20:10

Interesting, because I'm noticing as I've gotten into the space, even with the large language models, it's not completely attuned to women. Interesting. Yeah. It's a lot that can be done there. Yeah. Fascinating. So tell us a little bit about how you balance data privacy and interoperability while still delivering meaningful preventive recommendations.

20:39

Sure, I'm sure uh your listeners are familiar with companies like Cerner or Epic, which host HIPAA compliant data medical records. use uh Healthy, which is a similar company. They're a little bit more advanced with their technology. So we store everything on there and it's a HIPAA compliant AWS server. I mean, there's always a possibility of a breach. uh

21:10

but it hasn't happened with health records. I know we had that big breach in the healthcare billing system back in January of last year that created all kinds of havoc. So nothing's entirely secure, but whatever we do, whether we use healthy or build our own, it'll always be behind a HIPAA compliant firewall. Nice. You know, there's really a paucity of uh physicians, you know, throughout the world where there just aren't enough practitioners available.

21:39

Do you feel like, you know, with caution, AI can kind of be a little bit of a remedy for this? Like you talked about replacing your GP or being able to support your GP? Absolutely. And I say that AI is going to replace primary care. I think really it's going to replace kind of the onboarding and the monitoring. But whenever you need a procedure done, you're obviously going to go to a physician. But I do think AI will.

22:12

eventually prescribe medicine. And that's something we're working on to test in Singapore and Israel. Doctronics, one of our would-be competitors announced they're going to start doing a small test in Utah. They announced that yesterday. So, I'm sorry, and I'm not sure I'm answering your question. You're doing great. I mean, I think that, yeah, I think that it's- I remember. Yeah. So,

22:39

Physician in the loop. Absolutely. So importantly, we're building a platform that empowers consumers and uses as much data as the consumers can put in there and afford to put in there. Now, reimbursement needs to change for all these functional tests that are coming on the marketplace. Gut biome, RNA, DNA. I mean, there's so many tests you can do. And part of the things we're trying to do with AI is have that, or the doctor that we have in the loop now, but...

23:07

AI will replace it is saying, Hey, you know, Evan, what, is it? It's bothering you. What are you suffering from? What are you trying to achieve? And then we can determine which tests you should guess, get, and hopefully get them at wholesale pricing. But the problem is right now is that medical practice doesn't support a lot of this stuff. And when you go to your doctor and you have all these tests, it's kind of like walking into a ski shop and saying, look, I'm going up to Whistler this weekend and I I've got my, boots, my bindings and my jacket. bought on Amazon.

23:36

and could you mount my bindings for me? And oh, there's a zipper, zipper's broken. And the ski shop's like, well, you didn't buy any of this stuff here. It kind of pisses them off. And it's the same thing with the doctor. You go in with all these tests. First of they don't get paid to analyze it. It's not in the reimburse. That's the whole problem with the medical system. They're not paid for this stuff. That's going to change. But if you can walk into your doctor fully prepared with some pretty damn accurate data in your pocket, the physicians are gonna have to look at that stuff. And.

24:05

They'll want to the ones I talked to because like, I don't have to do all this stuff. Right. Uh huh. Well, here's the hope in, know, I mean, I think that a lot of physicians are burnt out, you know, and they don't want to do more work and they don't really want to do extra training. I mean, that's, that's why a lot of people come to see me is because they're like, yeah, my conventional doctor doesn't want to think about natural medicine or or integrative medicine or functional medicine or environmental medicine. Right. And so they kind of get, they of get stuck. So

24:32

I agree with you that this is necessarily a necessary evolution and that, you know, empowering the consumer is always going to be beneficial, right? So if they can have that knowledge, then they can be armed going in there. And I really hope that these physicians are going to meet them halfway and say, Hey, thanks so much for that information. And here's what I know to do with that information. I mean, look, look at your car. It's, less subjective than, um, or it's more objective than

25:01

health, is more subjective, but you've got to check engine warning light. There's a lot of electronics in your car and you know exactly what's wrong and when you need to go fix it and you know where to go. Right. So your mechanic is happy when you get in there because he knows exactly what to do. Right. Yeah. I mean, it's like, it's like the, it's the, it's the question about genetics, you know, which I've struggled with over the years. It's like genetics are usually about 10 to 20 % of somebody's health issue.

25:30

But it's like, okay, you get your genetics. Now what do you do with it? You know, it's one thing if you're looking at your telomeres, you know, from longevity standpoint, and you can see how long they are and how short they are. And if you do this intervention, then that increases the length. But like, what is the, like the genes don't change, you know, their expression changes, but the genes themselves don't change. And if you have that genetic information, how helpful is that going to be moving forward? What do you think? We think it's very helpful, but

25:57

one test doesn't fit all, we feel that you need some longitudinal data. And AI can now look across all your data. Again, we're still training the models to look at your genetics versus your blood versus your DEXA scan or whatever other tests, the more information is better. But doctors aren't gonna sit back and analyze that stuff. Right, right. Well, I mean, and I think if you, like you said, if you're providing the treatment plan to the consumer,

26:27

that ends up empowering them. can get a sense of that. And I think a lot of it, like you said, comes down to training. know, like, who's protocol? I mean, if you ask five doctors what their protocols are for a particular issue, they'll give you 10 protocols. Right. You what mean? You asked another question earlier that I didn't fully answer. came back to me. People are like, what if AI is wrong? What are the legal?

26:54

ramifications of that. And that's where the regulatory environment comes in. But I think consumers are smart. Now, sadly, a couple of weeks ago, there were two suicides associated with one of the LLMs recommended suicide as an option and two people took their own lives. I didn't see the press release. I just heard about it. I just don't see anybody saying, hey, AI told me to do this and it caused me more harm. Now, when we start prescribing medicine with AI,

27:23

there's going to have to be somebody that is liable. And I think that's where it's going to have government regulation that will put up a Chinese, put up a wall so that companies that are trying to help people with AI aren't going to lose their livelihood or businesses because AI made a mistake. Right. Yeah. I think that that's important. um And I also, I'm really curious to kind of see how the system ends up evolving when

27:52

You know, you could give me the same laboratory tests that you give another provider and we're going to have different recommendations. So then how do you choose who to go with? You know, like, or how do you train the AI based on, know, like, how do you consider, you know, you obviously want to train the AI on the best doctor possible, but how do you consider that? You know, like I'll give you a quick example. I'm just going to call my AI doctor here. putting on speakerphone.

28:22

How have things been going for you? Pretty good. Say hi to Dr. Hirsch. I'm on a live call right now podcast. Maybe we'll get him to do the program. Could you just quickly pull up my DEXA scan and just summarize for me the results? I'm still wanting to work on my visceral fat. I have started lifting weights.

28:46

It's great to hear from you, Hunter. And hello to Dr. Hirsch and your podcast listeners. I can certainly pull up your latest DEXA scan results for you. Your July 11th, 2025 DEXA scan showed a healthy BMI of 24.4 with a total body fat percentage of 24.7 % and your bone mineral density was within the normal range. It also noted an estimated visceral fat mass of 450 grams. It's excellent that you started lifting weights as that will definitely help with

29:16

body composition. We can discuss strategies to further optimize your visceral fat and bone density. Does that sound good? Sounds good. I'm going to jump back on this call. Just text me my latest workout plan. I'm joining a gym later today and I'm going to start hitting it hard. Thanks a lot.

29:36

Absolutely Hunter. I'll send over your latest workout plan via email right away so you'll have it for the gym today. It's fantastic that you're hitting it hard. We can always discuss it further on our next call. Have a great podcast.

29:50

So if this is very preliminary, if I go have another bone density test, it'll compare the last two and show me my results, how I've improved or haven't improved. It looks at everything, my blood tests, it develops plans. I just got a text from GSCAI with my plan. It emails it too. So it's very early, but you asked earlier about not gamifying it, but reminding. And I kind of want that. Not everybody wants it, but.

30:20

And you can adjust the settings, but like, good morning. Did you do your steps today? Did you work out today? How was it? How did it go at the gym? Eventually tie in all that data, you know, through like e-gym or all those wearable data that we can incorporate into now, you know, aura, sleep, Strava, and then build programs for people specific to what they want to do to, you know, exercise diet and whatnot. Nice. And so is this part of your longevity clinic?

30:50

This is the beta I just showed you. $100 a month plan right now. Yeah, it's in private beta. And there's a few spots open. You can put it in your poll notes if you want. It's ai.longgevityhealth.me. Excellent. Yeah, we'll put that in the show notes. This is exciting. I appreciate you coming on, Hunter. This has been really supportive, really interesting to kind of learn about where AI is going.

31:14

We'll drop that link as well as I think that there was another website that you wanted us to include as well. Which one was that? Longevityhealth.me. And the beta is ai.longevityhealth.me. Excellent. Well, it seems like an exciting future with AI partnering with healthcare and it sounds like you're at the front of it. So I really appreciate the work that you're doing on this because it's all for patients getting better, longevity,

31:43

you know, improved health outcomes and all that. thanks so much for the work you're doing. Thanks for coming on today. Thanks so much. So if you have chronic fatigue, whether it's from long COVID or chronic fatigue syndrome, go ahead and click the link below to watch my latest master class where I go deep into our four step process that has helped thousands of others resolve their symptoms naturally. After you watch that video, if you're interested in seeing if we're a good fit to work together, you can then get on a free call with me. All right, thanks so much. I'll see you over there.