The Potential for AI to Improve the Doctor-Patient Relationship - Morgan Cheatham, VP at Bessemer Venture Partners

Shiv Gaglani: Hi, I'm Shiv Gaglani. On today's Raise the Line, I'm really happy to welcome someone with whom I have something pretty special and unique in common: both of us are third-year medical students while still actively involved in healthcare business careers. 

My guest, Morgan Cheatham, is pursuing his degree at Brown University while also working

as a vice president at Bessemer Venture Partners -- one of the nation's oldest VC firms -- which has invested in major brands such as LinkedIn, Pinterest, and Yelp. At Bessemer, Morgan is leading investments in healthcare and life sciences. He’s alreay sourced a dozen investments and is the youngest investor to ever take a board director role at Bessemer.

It's perhaps not surprising that he was named to this year's Forbes 30 Under 30 list for venture capital because achieving major accomplishments at a young age has been a pattern for Morgan. By age twenty-one, he had created two new degree programs at Brown in neuroeconomics and in LGBTQ plus health that both continue to operate today. Before joining Bessemer, Morgan worked at Goldman Sachs in the consumer retail and healthcare group. 

So, Morgan, I've been looking forward to this conversation for a while. Thanks for joining us.

Morgan Cheatham: Thanks so much for having me.

Shiv: So, as we often start with our Raise the Line guests, we'd like to have you describe some of your career highlights because I know you started working as a data scientist at a healthcare startup at nineteen, so I feel like you're the Doogie Howser of healthcare investing and VC. So yeah, if you can just give us some of your highlights and what got you to med school in the first place.

Morgan: Sure thing. So, the headline for me is that there's always been a broad plan to be in medicine. But the way in which I've related to medicine has evolved over time. So growing up, I had a little bit of tunnel vision. I wanted to be a doctor from a very young age and in going through high school and applying to college, I just wanted to find the fastest path to do that. Like, what would be the most efficient way to become a doctor and start practicing medicine? And for me, it was applying to Brown where we have a joint BSMD program. The beauty of this program for folks who are unfamiliar is that you can participate in undergrad and your studies without the pressures of applying to medical school. You don't need to conform your resume to what a medical school might be looking for. 

So, coming to Brown, I was eyes wide open, exploring other areas. I had a very formative experience interning at a physician-founded startup called Kyruus, where at the time we were using large data sets to try to match patients to the right providers at the right time - a problem that the industry, as you appreciate, still hasn't solved today. It was there that I had this profound realization that data, software, analytics, AI -- whatever term of the season you want to use -- were going to have a transformative impact in the field, particularly in scaling this one-to-one patient provider relationship that I thought was the rest of my life -- that was what I was going to do every day -- and scaling that one-to-one relationship in a one-to-many fashion. 

So, that kind of became the guiding principle for the next series of steps that I took. After that, I realized that capital investment was actually another avenue for driving that nonlinear transformation in health care, so I spent a stint at Goldman where I learned that I was actually going to be a terrible banker, so that wouldn't be the path that I would take. It's a moment in my life where I actually took something off the table, which felt good. 

But then at that point, it kind of seemed too much and I just decided, you know, where I was at the time, I had all of these curiosities and interests and I wasn't ready to go to medical school. So, I joined Bessemer Venture Partners really out of luck. I stumbled into meeting a partner who was on campus at Brown one day, asked him if they were hiring and three weeks later, I had deferred medical school and decided to join the firm. Fast forward, I took a two-year deferral from medical school. That two-year deferral quickly became four because I absolutely fell in love with what I was doing at Bessemer, the founders I was able to meet and work with and everything that I was learning. 

One stat that I love about working in venture as an analyst early on in your career is you meet

anywhere from 500 to 1,000 companies during that time frame and so in many ways, you're learning business in the wild through these experiences that you're having with companies. Long, long story short -- and I'll wrap it here...happy to dive deeper into any other parts of the story -- I had this realization during the pandemic that, you know, I wasn't getting any younger and I was also recognizing in the founders I was speaking to that there were really interesting things happening at this intersection of biology, computation, medicine and delivery. And in my existing posture, I wasn't really seeing a lot of parts of the industry that I desired to see and that was really on the delivery side, on the research side. 

So, I made the challenging decision to go back to medical school. I moved from New York to Providence and I've been here ever since. And now, just like you, I’m a third year and that kind of catches us up to today. 

Shiv: Yeah, it's pretty incredible and I know a story that a lot of people are interested in because we're both part of this “MD-plus” community that our friend, Sherman Long -- who's also a VC investor -- helped set up when he was at, I think, Mount Sinai for med school. There are so many clinicians, medical students, pre-meds even, who I'm sure you interact with pretty regularly, I do as well, who are asking about these careers: either they want to do health care tech as a founder, or as a VC. I'd love to get your opinion on what are some of the things that make you a good venture capitalist based on your drive as a medical student or maybe weaknesses that you had to compensate for? 

The reason I ask this is that when I left med school the first time to start Osmosis, I wrote this article for Entrepreneur called Why Med Schools Are Pumping Out Entrepreneurs, because there are a lot of shared similarities between founders and med students like grit and running a marathon versus a sprint. But I'd love your thoughts comparing and contrasting VC with med school.

Morgan: Funny enough, I actually think there's a lot of similarities between the two fields, even though when you think about your average college student going into venture and your average medical student, those two people might seem at odds with one another. On the venture side, we're really in the business of forecasting the future, so constantly trying to figure out where the puck is heading in technology and that requires an underlying desire to learn and kind of an insatiable curiosity, and that's personally something that I've lived with, struggled with, tried to embrace kind of my entire life. 

Then on the medicine side, as you appreciate your first day of medical school, you're sat in a chair and told, hey, welcome to a field of lifelong learning. It's not going to stop. You know, the studying won't stop, the exams won't stop and the constant kind of refreshing your current understanding of pathophysiology and the human condition is going to evolve over the arc of your career. So, curiosity is this common theme that I find to be really interesting in both of these industries. 

The other I would say is ambiguity and learning how to be comfortable with ambiguity. In venture, when we're assessing a company and examining its performance, it's not like we look at a checklist and we say, “OK, the company has X, Y and Z, therefore we make the investment.” There's much more gray area, even down to like, the CEO and understanding their strengths and areas of development and understanding where the company's core kind of edge will come from in the long term. 

And in medicine, as you appreciate, as much as we like to believe that medicine is data driven and consistent and reliable, you're not even able to replicate many of the studies that are in current scientific literature, and oftentimes no two cases of the same disease look the same. And so I find that comfort or even an interest in navigating situations with ambiguity is a common feature of folks that straddle both these worlds, and is certainly something that I find an interesting challenge day to day.

Shiv: That's awesome. I'm sure if you haven't written an article about it yet, I think there'd be an interesting collaboration between you, Sherman, and maybe Robert Lord at LionBird, who's also a mutual friend. You produce a lot of really good content. I really enjoy whenever you post something interesting, whether it's, you know, hiring checklists for companies and how do you assess the biotech industry? One piece of content that really put you on the map, and I think I reached out after this, was your USMLE paper with ChatGPT back in December with Dr. Nigam Shah who you introduced me to and we also recently had on the podcast, so thank you for that. 

I would love to just hear whatever you're willing to share about that paper because you live in this world of technology. For our audience that doesn't know that paper, tell us about how you got involved in that paper, and then as a med student right now, how are you using technology to maybe alleviate the burden of being a med student?

Morgan: Yeah, so taking a step back, my interest in computation, AI and data and health care really dated back to that first experience I had as a data scientist working in the wild on these problems. As I joined the venture community -- kudos to my colleagues here at Bessemer -- but in 2018, we created a vehicle for investing in early-stage health care and life sciences, AI companies that was called our Deep Health Seed Fund. 

Now, clearly, we're not in the business of marketing. I couldn't think of a better name. But essentially, this was an acknowledgement, again back in 2018, that computation and biomedicine were going to intersect and produce some of the world's most valuable companies, and we still believe this to be true today. It's only kind of garnered more interest and support from the ecosystem. So, I want to just kind of clarify, it's been an enduring interest for me, but it's been largely like a practical interest and something I've done professionally. 

Going back to medical school was a really unique opportunity to dive more deeply into some of these areas from a research perspective and to do so kind of free of the constraints or

the pressures of business or finance or commercial interests, frankly, and has allowed me to really intellectually explore the research questions that I find most salient. So, when GPT came out -- specifically in the form of ChatGPT that most people became aware of because it was accessible via this chat like user interface -- that to me was a really unique moment in medicine because, one, I started using it to study and I thought that was like a really interesting opportunity. I think I probably passed Step One because I was using GPT to quickly reference answers and the like. But, two, I actually saw colleagues pulling up in the hospital and saw the way that access to information in this specific way -- this this kind of command line interface for medicine that I've also written a lot about -- was going to be transformative. 

So, I thought no better opportunity than to kind of marry some of these research questions with the leaders in the field who are asking them -- and that's Dr. Shah at Stanford -- to say what will be the clinical utility of large language model information retrieval at the point of care? That's the very kind of initial question we set out to answer. So, just kind of summarizing some of the work for folks listening, we essentially gathered a number of questions from physicians at the point of care at Stanford. Oncologists at Stanford had different questions about how they might proceed with treatment or a cardiologist had a question about a particular intervention. We gathered all of those questions and then we fed them through a large language model, specifically we fed it through ChatGPT. 

What we were interested in analyzing was, what is the value of this information at the point of care and before even getting to value, is the information that a large language model is able to synthesize safe? Does it potentially cause harm? And if it causes harm, what are the implications of that, right? All of these questions are being asked in the spirit of what would it look like to have AI specialist consults in a real-world environment. So, that was kind of the initial question we started to answer. 

You know, derivative work that the lab is now working on has really started to say, OK, we have large language models that are being applied to various clinical tasks...how do we build the right validation benchmarks and evaluation criteria to know whether or not the task is performing well? Obviously, we can always ask another physician to say, was this answer safe? Was it harmful? Was it clinically useful? But if we're going to scale the kind of research that we're doing and really understand the implications at a high level, developing these standards will be critical for the industry, and that's across clinical tasks and biochemical tasks. So, that's kind of where the future direction of the work is heading. 

But as you can tell, it was like super applicable to my day to day being able to use ChatGPT and then kind of ask questions about the utility of these technologies directly at the point of care.

Shiv: Yeah, absolutely. So, so exciting because I mean, things are improving so quickly. We recently also had Michael Howell, who's the chief clinical officer at Google, on the podcast, and he cited that the performance of MedPaLM on USMLE went from 60% or lower in December to now over 80% in May this year. So, who knows...by the time we're actually in practice -- if you choose to go that route -- how good these LLMs will be with GPT 5, 6, 7, etc. 

So, how has it been returning to med school? Has anything surprised you? Are you getting a lot of questions from fellow med students about your time in venture? Are you still being pitched? How are you balancing being a board member while being in med school? I would love just your overview of the return to med school.

Morgan: Yeah, I imagine it's one you can relate to, but it's certainly been an adjustment, you know, to go from having full autonomy of your schedule as a venture capitalist -- which is one of the best perks of the job is that it's a very autonomous, creative job -- to being told where to be, when to be there, how long you're going to be there, when you're going home. A lot of that has been a big adjustment, candidly.  

But I actually I try to keep the two worlds separate. That sounds funny and strange. Like, why wouldn't you integrate the venture work more in the medical side? But like, I'm here to learn and train as a physician, I'm not here to pitch products or source the next great company. If I just wanted to do that, I could have stayed in venture. So, frankly, I've kept the two worlds largely separate. 

I have wonderful classmates who have interests across all areas of medicine, from evaluating social and structural determinants of health, to some of these deeper questions in molecular biology that I've alluded to which are of interest to me as well. If folks have questions about past lives, I certainly answer it. I think what's really special about medical school is you realize that there are a lot of people who have past lives. We have folks who were previously veterans serving our country. We have folks who have started and sold companies themselves, folks who are doing research, who have other advanced degrees in law and PhDs and MPPs. So, I don't really feel that different. Like, I have my own story, but I've approached medical school with this belief that I have something unique to learn from everyone who's here and I’ve really tried to do that in my time so far.

Shiv: That's awesome. That's really good. And I think as people increasingly take gap years or enter med school non-traditionally -- not right after pre-med -- I think that that's a really good perspective to adopt. I would encourage people to do the same in whatever they're doing, even if they're not going to med school, but if they're joining a new company. You can learn from anyone at any point. 

What have you been enjoying in med school and are you thinking of doing residency? What are you thinking after med school?

Morgan: I knew you would ask this question. You know, I'm honestly right now, like you, I'm doing rotations and it's day-in-and-day-out in the hospital, in the clinic, seeing patients, trying to get as many reps in and experiences in as possible. I'm really focused on learning what I can from these experiences. What's been most striking to me, frankly -- like, pre-Morgan being in med school to now being in med school -- is I think I had very broad assumptions about how black and white medicine was, and as I alluded to earlier, it's so much more. I think today it's both an art and a science -- and that can be uncomfortable at times -- but I think it’s where a lot of the opportunity lies when we think about how the field is going to transform over the next ten, twenty, thirty years, and what I specifically mean is our ability to use large, multimodal data sets and computational technology to better phenotype and characterize disease, for one. 

Like, a lot of the diseases that we diagnose today are actually tens, dozens, hundreds of diseases that we're simply classifying all as one and treating with a monolithic therapeutic regimen. And so I'm really excited about a lot of the computational phenotyping work that's being done and I actually just joined a lab called the Zitnik Lab at Harvard that's focused on some of these questions. And so I think for me, as I continue to rotate, if I fall in love with something like a particular specialty -- internal medicine, OB-GYN, et cetera -- it's certainly not off the table. But I'm also really keen to run down some of these questions I have at this intersection of AI and biomedicine and may, I don't know, may pursue a degree that's focused on those areas like a master's or a PhD as well. So, keeping the options open, but honestly, just trying to learn as much as I can.

Shiv: Yeah, that's awesome and I think it's the parallax of choice that people who are naturally curious like yourself and me face where, you know, it's hard to get really good at something if you're interested by a ton of different things, but that’s actually where a lot of value is created. If you have a deep enough experience or interest level in multiple areas and then find those creative ways to combine them or find people to work with who are really deeply experienced on that particular topic. That could be a good transition actually to some of your investments. I'd love to hear about your time at Bessemer and even now, what investments kind of pop up? I mean, I know I'm not going to ask you to choose your favorite child here, but...

Morgan: There are no favorite children in venture. 

Shiv: Exactly. No favorite portfolio companies. It is a power law business. I assume there's some favorites, but I'd love to hear what comes to mind as far as companies you've invested in? I mentioned Abridge, so maybe you can refresh our audience on that one, too, because I think we're both really excited about that one.

Morgan: Yeah, happy to talk about Abridge. Maybe just to give folks a survey...my work at Bessemer is super broad. We'll invest in anything from health care delivery businesses -- so companies that are actually providing services, both brick and mortar and virtual. Some of the investments I've worked on in that space have included Hinge Health in the musculoskeletal care space. Ginger, now Headspace, in mental and behavioral health. So, a large focus on the payer-provider world. 

And then I'd say, also in the biomedical kind of AI realm, I've worked with a company called Subtle Medical that spun out of Stanford and is using deep learning to accelerate medical imaging. So, taking that ninety-minute MRI that maybe one of our listeners has had and shortening it down to fifteen minutes by taking low-quality,  low-resolution images and using denoising algorithms to bring them to clinical diagnostic read quality. A number of really exciting applications there also in terms of lowering gadolinium or contrast dose for imaging and being able to use kind of algorithmic contrast to enhance imaging. So, clearly relevant for patients with renal conditions and also in children. 

And then Abridge, which you mentioned, is I think relevant for this podcast as a physician focused startup and also physician founded startup. Abridge essentially transforms the physician patient dynamic conversation by allowing doctors to capture all of the rich information and stories that patients share at the point of care into the medical record ambiently without having to sit in front of their computer and type it all out themselves. What's so exciting about the company is they're using a lot of the technology that we've been talking about -- large language models, generative AI -- to capture and transcribe this information that patients are providing every day and structure it in ways that are useful for downstream tasks. 

One such example that you and I know well is writing the medical note that drives a lot of the downstream billing tasks that hospitals rely on to keep the lights on, but also generating other documentation like summaries for patients at appropriate reading levels or in their native language so that they understand what happened and can share what the doctor said or what was learned in the visit with loved ones and caretakers. 

I think what's so special is -- as I talked about earlier in our conversation -- a guiding light in my career has been how we take this sacred one-to-one relationship and scale it one to many, and Abridge is the company that does just that. It takes the sacred patient provider conversation and empowers doctors to do so many other things when that's captured correctly and without additional burden on their part. So, very excited about that company. 

I actually find that I use Abridge every day as a medical student. As a funny story, the other day, about a month ago, I had to take an OSCE -- which, for folks who aren't familiar is a structured kind of clinical exam that we take as medical students with patient actors -- and I was abruptly reminded of the reality of medicine when I was told I wasn't going to be able to use Abridge to take my notes in my visit with the OSCE actors and had to resort back to pen and paper. And anyone in medicine who's had to make that transition back...like, writing on pen and paper when you're wearing gloves and have PPE on like the whole thing. I mean, it's a total mess. So in that moment, it clicked for me like, you know, Abridge is the future, and it's one of the companies I'm really excited about.

Shiv: That's awesome. Actually, since you mentioned you use Abridge for patient notes, what other tools or apps do you use as you go through med school? Because I think I would love to learn from you.

Morgan: Well, I have to say I use Osmosis for learning. So, that's obviously on my list. You know, there's some really cool up and coming tools. Glass Health is a tool that I really like. It was founded by Derek and Graham, a former Brigham resident, actually, in internal medicine and a seasoned digital health engineer and together they've created kind of a Notion like platform for doctors to be able to keep track of all of their medical knowledge and understanding as they're going through training and beyond. 

We talked about how medicine is really an industry and a field of lifelong learning and oftentimes we're keeping track of the things we learn from a particular case or from a particular lecture on scratch paper, or on the back of napkins. I've seen people writing on the back of masks. I mean, I've seen people use trash to take notes in the hospital. It's crazy what people use. So, what Glass is doing is it's acting as this central repository and really a knowledge graph for clinicians to understand where their knowledge is robust, where there's opportunity to learn more and how it evolves over time. I use Glass to make sure that I'm covering all the bases in any particular area.

I would be remiss not to mention MDCalc, which has been around for just under twenty years, but is used almost universally by physicians for medical calculators; and also platforms like UpToDate, which we've talked about. I think many of us are kind of living in the shadow of Dr. Bud Rose, rest in peace, who really paved the path for many of us to be a physician informaticist and to think that's a really cool and exciting thing. So, those are a few things, but you could probably fill in the rest given some of your experiences, too. 

Shiv: Yeah, that's a great list. And MDCalc in particular. I don't know if you know Dr. Joe Habboushe who is a friend. We had him on the pod as well. He's a great guy. And yeah, he's done a lot with bootstrapping to get to this point. I know we're coming up on time, so I only had a couple of last minute questions. 

First, just what advice do you give to, say, younger premeds or other peers of yours about approaching their careers?

Morgan: Yeah, so I'm a big believer in Bayesian statistics, and I've talked about this a bit in other areas of my life. For those who aren't familiar, Bayesian statistics is this field in statistics that really focuses on hypothesis generation and your ability to predict future events based on priors or kind of past understanding, and at the highest level, that's how I approach my life. I'm constantly training my priors. When I have to make a decision or I have to do something, I'm sampling from those distributions that I've trained. 

So, when I think about this framework for life and for learning, it's all about exposing yourself to as many things as you can and as many diverse perspectives as possible. I had a professor in a graphic design class I took when I was at Brown who told me that our job at this point in our lives was to consume as much content as possible, and I think that's kind of like a corollary to this Bayesian framework I've laid out for myself, which is just trying to do that. 

As you see in my life, I'm consuming content from the business world, the clinical side, consuming my own content kind of firsthand from the patients and really hearing their stories and understanding the softer sides of medicine. And so that would be my best advice is practice like a Bayesian ideology, train your priors, and I think you'll find yourself in some really interesting positions that maybe you hadn't planned for, but there's a reason that you end up there.

Shiv: I love that. I'm a huge fan of that kind of approach and different algorithms to live by. One of which I often share with people who ask me for advice -- and I think one you've probably heard or given yourself that's related - is the “explore versus exploit” algorithm. The analogy often used to say you're like, living in a tribe, right? And there's fertile ground here where you can farm, you can raise sheep, whatever. So, you can exploit what you know works, this fertile ground works. But then a certain percentage of time, you may want to either have some people in that tribe, or yourself, explore other pastures...maybe climb that mountain and see what's on the other side of it. Could be a river with more fertile ground, more sheep or whatever it is. I think a lot of my career has gone through these periods as well of going very exploratory, like, two or three years of just consuming as much as I can, producing a lot of things. When Osmosis was created, I created like three other things at the same time. 

Morgan: Of course you did. Naturally. 

Shiv: You too. I mean, I'm sure you have that, too. But then once Osmosis was working, focus and explore that. I think a lot of people like us have a lot of trouble with that. When do you switch? Because for middle school, high school, college and even med school, you know, just through sheer brute force, just working really hard or mental capacity or whatever, you can keep exploring and keep doing a bunch of things really well, but nothing really that focused or exploited. You know, exploit sounds bad because it has a connotation, but I found that model going between periods of exploration and exploitation to be really helpful. 

Morgan: I think it's a great framework, and I would say my biggest regrets in life so far -- when I think back on what those are specifically -- it's always been opportunities, to use your analogy, where I didn't exploit, where like I thought something was interesting, but I didn't ask the next level of questioning or I didn't pursue the meeting with the person who had written the paper, done the work or started the company. I just said, “Oh, that's interesting” and left it there.

But if I had just gone that extra step and gone a level deeper and maybe, frankly, cut out other things in order to do that, I think I would have found a lot of value in that. So, I completely agree. I think careers wax and wane in terms of breadth and depth and it's our job to kind of be the arbiters of when the right time is to make each of those hard cuts in each of those decisions. So, that certainly resonates with me.

Shiv: Totally. Awesome. Well, last two questions. One is just whenever I meet smart people who have done amazing things like yourself, I like to ask them about any influential books or podcasts you recommend, things that our listeners should do, including myself.

Morgan: There are a ton and they come from all over, like, they don't just come from medicine or science, frankly. I'm a huge classics nerd, and actually I originally wanted to go into college to study classics, and my dad told me he wouldn't pay -- I hope he doesn't listen to this -- but he told me he wouldn't pay if I studied classics. I did science and classics on the side. So, a lot of the things that inspire me are from mythology and my work translating Latin. I'm a huge fan of Ovid and Virgil and Catullus for fun. 

I would say more recently -- and what might be relatable because a few people listening to this will probably want to go back and dust off their old Latin books -- I love Judea Pearl's The Book of Why. I don't know if you're familiar with that one, but he's a prominent computer scientist, and I think the book kind of outlined some of the frameworks that we're talking about in terms of ways to view the world with an inquisitive lens. I've read it a few times now, and every time I take something different back from that. 

Then another thing I do for fun is I read -- and this relates to my venture work -- but I love reading old school business books from like the 90s and 2000s... just hearing about some of those stories I find interesting. There's a book called The Gorilla Game, which is all about how to pick winners in high technology or something to this effect. There's some great principles in there for what to look for in terms of understanding whether something is a true paradigm shift or a moderate improvement in society and quality of life.

I relate this to what we're seeing in AI where there's a lot of opportunities to deploy AI to make something like, ten or twenty percent better... you know, moderately improve or streamline a process. But in venture capital, we're playing for like, massive outcomes. Ten to twenty percent better is not good enough. And I think what so excites me about this era that we're in --just to tie back the theme - is that the technology and the AI we have now is actually enabling novel behavior, and that's kind of a criteria I've now created, inspired by this book, of what new behaviors are companies now enabling and empowering? I think that's a really interesting heuristic for understanding whether something's going to be transformational, venture scale, or maybe better suited for another asset class. 

Shiv: I love that. I wouldn't expect anything less. I'm glad you had such an eclectic variety there. And I agree with you. Reading has been the biggest unlock -- both as a founder and as a human -- because, as I think Ryan Holiday said, you get wiser without getting older when you can just tack on other people's years and years and years of history and it's nice and dense for you to consume.

Morgan: If I could add one thing to that, sorry, just really briefly, I also think it's important for folks to not worry about where they're finding inspiration or sources of new ideas. Like, I don't believe that books have a monopoly on good ideas. And so for me, also, a lot of short essays, a lot of, frankly, like Twitter tutorials or Twitter storms that people put out have been some of the most impactful on my mental models. Tick-Tock... you name it. In this consumptive methodology of training your priors, I think you have to be open to a lot of different forms of media. So, that's just another thing I try to keep in mind.

Shiv: I feel like we just have like another whole podcast on how we think about this consumption. One thing I would like to add to that, though, is one trap I feel I've fallen into -- especially of late during these explore times -- is consuming too much content, being too much of a consumer. Consuming helps with the priors, as you mentioned, and exploration, but sometimes, like, the best source of adapting my priors has been action. When I finally take action on something, that's where I've learned the best, where I remember things the best because I've actually run a P&L or placed an IV. When I take action on something that's far more powerful than just consuming. So, it's like even that there's like an explore versus exploit of how much to consume versus to create or to action.

Morgan: Couldn't agree more.

Shiv: Cool. All right. Yeah, I felt like we would both agree on a lot of this stuff, but we should have a debate sometime because I'd love to see where we disagree on things because that's probably where we're at.

Morgan: Let's find it.

Shiv: Yeah. Well, OK, so my last question is an open mic. Anything else you want to share with our audience that I haven't asked you about yet?

Morgan: Yeah, I mean, I just say for folks listening, if anything I've said has resonated with you or if you've disagreed strongly with me on anything, I'd love to hear from you. On the Bessemer side, I can be reached online and happy to link some of the ways to contact me. But if you're thinking of building in biomedicine or AI or you're passionate about some of these areas from a research perspective, I'd love to chat. I'm always available.

Shiv: Awesome, Morgan. Well, thank you so much for taking the time to do this on the podcast. It's just a real treat to get to know you and follow your career journey along the way. I'm excited. Wish you the best as you finish up med school. 

Morgan: Thank you to you as well. It's been an honor to be here and I hope we can cross paths on an away rotation and in the near future.

Shiv: That'd be awesome. Well, with that, I'm Shiv Gaglani. Thank you to our audience for checking out today's show and remember to do your part to raise line and strengthen the health care system. We're all in this together. Take care.