E20: AI Agents & The Intelligence Age: Hype vs. Reality

In this episode, the hosts discuss the current state of OpenAI, including recent departures of key personnel and the implications for the company's future. They explore investor perspectives on OpenAI's valuation and growth, as well as the challenges faced by high-growth companies. The conversation shifts to the role of AI in healthcare, highlighting both the potential benefits and the regulatory hurdles that startups face in this sector. The hosts also discuss market entry strategies for healthcare technology and conclude with thoughts on the future of AI in various industries.

Takeaways

OpenAI is experiencing significant growth but faces internal challenges.
Investor sentiment is mixed regarding OpenAI's high valuation.
Departures of key personnel can indicate deeper organizational issues.
Product readiness is crucial for maintaining market trust.
Healthcare AI presents both opportunities and regulatory challenges.
Startups must navigate a complex landscape to succeed in healthcare.
Partnerships with established players can ease market entry.
AI has the potential to revolutionize healthcare delivery.
Long sales cycles are a common hurdle in healthcare tech.
The future of AI is promising but requires careful management.

Links Discussed

Episode Transcript

Introduction

Rod: Welcome to another episode of the Chris Rod Max show. This is episode 21 - we're already on our 21st episode! As seen every week, I'm joined by my co-hosts. Hi Chris!

Chris: Hello everyone, good to be back.

Rod: And Max.

Max: Hello, good to be back.

Rod: Every week, something is always happening in AI. What we do in this show is try to put this in perspective. We discuss what this means for those who are adopting technology, rolling it out, and want to maximize its use in industry and business.

OpenAI: Recent Developments and Valuation

Rod: Pretty much every company, almost all Fortune 500 companies, are working with OpenAI. The past weeks have been full of discussions, news, debates, and so on about what's going on in OpenAI. We have on our screen now a tweet from Gary Marcus, a so-called AI leading critic, where he's talking about all the departures that have happened in the last weeks. Plus, he questions if this justifies the evaluation of the company. So Chris, you've been reading about it as well. You have some opinions. Maybe you can put in perspective what is happening at OpenAI right now.

Chris: I think a lot of things are happening, and actually, it's not something that very much surprises me. I mean, I think what we see is a hyper-growth company. And actually, I have to say it's not a company yet - it's a nonprofit that is growing exponentially. Everyone is super, super interested in working with OpenAI, building new customized software for their own use cases.

And obviously, what we see is that the organization is a little bit lagging behind. And I think I can only imagine that inside OpenAI, it's sort of like a nightmare of restructuring, people coming in and out. And it's just a lot of dynamics. And I think that's also something that I wouldn't say is surprising given their growth, given their attention, given their pressure from society and the outside world.

It actually very much reminds me of perhaps companies like WeWork that also had a crazy growth curve. I think internally there were a lot of things to deal with. And as someone who's grown companies, I can only imagine how tough it must be to grow exponentially also within your team.

There's a lot of things you need to figure out from processes, incentives. And I think also the latest news on OpenAI and many, many people leaving - I can also understand because there's probably a lot of frustration. People come into a company because they believe in a certain purpose. And I think especially with OpenAI being a nonprofit, being sort of this like cozy little geeky research lab that wants to do good for the world, is now faced in a position where there's a lot of commercial interest, opportunities, and obviously a big decision to make whether or not to make the company now a real company and not a non-profit.

And hence, obviously a lot of people not agreeing with that or not wanting to support this kind of purpose.

OpenAI's Valuation and Investment Potential

Rod: Yes, and here you present many, many topics that we can digest and try to look at individually. So let's start with the departures. We're not talking about some interns who have left, but we're talking about the CTO, the lead scientists, the president who is on leave. And at the same time, the company is trying to raise around $5 billion. Some people are talking about$ 6 billion to value the company at $150 billion.

Max, from the investors' perspective, when you have a company that on one side has this super strong brand, also that has so much potential, but at the same time you see it seems so many areas of the business are burning, their top talent is leaving and so on. How can you decide or what is your framework to think about? Should I join this round or maybe I wait until things calm down, but potentially miss the opportunity of joining now the set of investors at OpenAI?

Max: I think a true story here - I don't know how much you guys follow on the angel round. So basically, a lot of folks get an opportunity to invest in OpenAI. But on the private rounds where they're doing secondaries, I remember seeing it at like 30-something billion, which was the last round price. I actually got to see the term sheet. And at that point in my head, I was like, wow, this company is growing like crazy. 30 billion, you didn't invest. And now it's a year later, it's about five times the price. So it's 150 billion.

But I guess, if I were to put an investor hat on, the question becomes: if we were to use the traditional valuation method and try to think about this, right? So they are trying to hit 3.4 billion revenue compared to 2 billion last year. And the funding is now pushing up extra 5 billion, 20 billion, valuing them at 150 billion, which means you could basically say that they are like 75 times if it does something along that line, which is really, really high.

If I were to think from that perspective, it feels high. But then if you look at their growth, it's actually quite mad. 2022, their revenue was about 200 million. 2023 is 2 billion. So the way they grow is a bit mad. I don't think you have seen any companies like this before.

So the way you would think about it is whether or not you think of it as a traditional company or you really think of them as hyper-growth. Throw away all the framework that you know from value investing and really go for the ride. Go for the growth. It feels like a lot of the investors that are investing in it feel like this right at the moment.

Potential Risks and Investment Considerations

Max: My question is what happened to the early investors, like all the Khosla ventures that did the very early round? Are they still following rounds like this, or are they just keeping the amount of equity they've had so prorated so far?

I think the new investors coming in, we have seen this with larger rounds. Often than not, later rounds investors, some of them do lose money because the valuation is a bit mad. But for OpenAI, I must say, I still believe in some justification of why that is.

Instead of looking at why it will become so much, perhaps look at what are the potential risks that come with it. When I say risks, how would OpenAI eventually just die? It will be a combination of some sort of large tech incumbent coming in. So you think of Meta giving out for free, entirely upending their business model. You think of some sort of Google or Meta forcing their own AI into their gateway to the internet, right?

So like the way we would go to the internet is via all these Meta and Google's applications. So if they were to force AI onto those applications, does that then create, I would say, forces OpenAI to play on the enterprise side? I guess those are the questions that I have in my mind.

What's the downside? What will cause this $150 billion to evaporate tomorrow? At this point, I still don't think it's super likely, just from the growth they have had, the training they have gone through. Yes, there are a lot of senior departures, but I agree with Chris. It's probably more purpose alignment because a lot of folks that joined OpenAI initially was to do good and then nonprofit.

So I think we have to take things into context, I suppose. That's what I'm trying to say with all these revenue numbers, as well as the departure, etc.

OpenAI's Potential for Profitability

Chris: Maybe a quick question to Max. I mean, do you believe they will eventually break even? Because I think that's really the question - whether or not they can be profitable. Otherwise, what would be an exit strategy for OpenAI in your opinion?

Max: So very good question. They say they expect to lose roughly 5 billion this year. And their revenue is expected to reach 11 billion. If they do hit that, and the question is how much are they going to lose next year, they are likely going to get breakeven, right? Even just by numbers that you're looking at now.

So they already hit 3.4 in May 2024. So how much more revenue are they going to get? 1.6 billion for the rest of the year in order to meet the 5 billion mark? Probably. But I have a sense that they are actually raising to grow even faster. They will need a lot of capital to train the models.

And at the moment, remember, their largest cost is also their largest investor, which is Microsoft. So I think they have the ability to actually break even if they want to. But that would still the growth that early joiners leave a company that there is some revolving door early days.

Organizational Growth and Transformation

Rod: So in an organization, the organization grows, let's say, standard in the company, right? So we know that the profiles that early joiners have when the company is small is very different from the profiles required when we're talking about a large organization valued in billions of dollars, having billions of revenue and so on. And this is very common. People feel that they do not fit, or maybe the organization thinks they do not fit anymore and change happens.

Nevertheless, here we're seeing that this is happening constantly. So, potentially this might be something related to culture, and as well as the fact of what Chris was describing on changing mission.

So Chris, when you see these types of teams that are having these transformations, what do you advise them or which frameworks do we have available in order to, on one side, let's say, enable this change, maybe bring more senior executives, maybe bring someone who has experience in managing billion-dollar companies and so on? While at the same time keeping the innovation, also retaining some talent because I can imagine that now these are like the names that we're seeing because they're famous. But behind the doors, there might be others who are also following suit, who might be joining those who are departing now.

Chris: Yeah, right. I don't think there's a silver lining or like a framework here, but I think it's really about looking at the maturity curve of a company, right? And when you start out, most likely you have two types of people. You have the founders, the visionaries, these entrepreneurs that are very pragmatic and trying to just crack the problem. And then you have on the other hand like operational people to really execute, right? Normally it would be engineers and product people as well as like salespeople.

That's sort of like your early journey, your 10 to 15 people kind of frame. And then when you really find this product-market fit and you really have this crazy sudden growth, that's where chaos happens, right? Normally. And then throw in another geographic expansion. You have to deal with like multiple offices, etc.

And that's really where, you know, organizational like professionalism obviously comes into play and you have to think about all these other things like communication channels, decision-making matrices. Also even just like keeping people happy, right? Like I think at the beginning it's really, you know, the incentive for people to join a company is the vision. It's kind of like feeling important to contribute to a very important mission.

But then obviously, people think about their own careers. They want some career progression. They want to know when they get the next salary increase, etc. And then you have to like design all these like frameworks and these standards and these policies. And all of a sudden, you are in this like mid-level of company growth where, let's say it's series B or C or depending, you really need to think about, OK, what's the kind of organization you want to become?

And usually the visionary entrepreneurs really hate these kind of environments because it's, you know, it feels a bit more corporate-y. You have a lot of policies and standards you need to adhere yourself to. Security might be even a thing that, you know, pops up, compliance, all these kinds of things. And that's really where you also see lots of people transitioning out their management or the investors asking for a more professionalized CEO to really manage the organization.

There's also the phase of a company where you normally have like a cash cow, like a core business, like something that you actually want to only incrementally improve and actually want to protect and maintain. And I think that's sort of like how companies work.

And this is also one of the challenges I think bigger companies or corporates are facing where, okay, it's just about kind of protecting this core business. And that is also the reason why it's so hard to then reinvent yourself and come up with a completely different core business, because you don't have the right people in the organization. You don't have the crazy innovators, the chaotic ones, but you actually have the managers, the operators that are just trying to make sure that things don't crash and that revenue and EBIT incrementally grow and grow healthy.

OpenAI's Relentless Product Development

Rod: Yes, nevertheless, the company remains relentless where although we have all these departures happening in parallel, the company hasn't stopped. They keep releasing new products, they keep improving their existing ones. We see that just this week there was a new announcement of new functionality, new features, and we're seeing how the product matrix of OpenAI is widening.

So on one side, of course we have new models coming all the time, but also they're adding functionality on top of these models, such as caching the prompts to reduce costs, but also enabling the development of smaller models from larger models and so on. So it is becoming a very large portfolio matrix.

And on one side, we have this situation where this way they can pretty much occupy every niche in the market, but also every new product means managers, means those who have to maintain the code, etc.

Max, when you see companies that have a large portfolio suite, what are your thoughts? Do you think this is a positive aspect because you think, hey, this way they can really attack every niche in the market? They can really have individual pricing structures, have every segment available, or do you say, hey, this will lead to a lack of focus? In the end, you might do a lot of things, but maybe many of them might not be that great and so on. So you should identify your flagships and then bet on them.

Max: It's a key question for large corporates. Like how do I think about it? So in the market, let's say if you were to do an investment, number one, a lot of, I would say, investors will price in a monoline versus a multiline company just as a way to diversify risk in the sense that you have different revenue streams. So there is actually a premium to a multiline, multi-line company.

And I must say though, from an OpenAI perspective, I don't think their products are so different from their initial product. So they're actually not straying very far away from what they're doing as a core, which is foundational LLM models. Everything they're building is almost, I would say, adjacent to what it started as, as the core of building AI.

So I must say that's number one. If they come up with more features, finding more ways to capture value and providing value to the end user, that's actually a good thing. It's just, you know, they have to remember what they're doing on the core perspective.

It's a little bit like, you know, I don't know, American Express started going to do more restaurants and less on their payment network. Then that will be a little bit of like, OK, what are we trying to do? If American Express started opening up new restaurants themselves, then that would be something that you would think about.

I think for now, from OpenAI's new product, especially on their roadmap specifically, I feel like they are still not too far away from what they are trying to do.

Because I was just looking at all the differences. So at the moment we're saying OpenAI is at $150 billion of market cap. So if I were to look at top 100 market cap, that would place them at$ 105 just above Commonwealth Bank in Australia. And then everyone above Uber, Total Technologies, Lowell, Goldman Sachs, all of them have multiple products, right? If you were to go all the way up to Apple.

So in the sense that, you know, I don't think they're doing something too wrong. But my question is, you know, if you have the money to invest as normally you do as an investor, where are you planning to place your bets? And nobody would say none of the companies above 105, 104 and above has the same growth rate as OpenAI.

So that's probably why contributing to a lot of the crazy valuation and crazy growth. And if I were to come back to the numbers on the product itself, we're seeing that they hit about 3.4 billion. I think one of the reports says that they are about to hit about 11 billion of revenue next year.

So if you compare 11 billion to 150 billion of evaluation, that's not too crazy. The question then becomes, will they hit the 11 billion? And then how much cost is associated with the 11 billion?

So I guess I remember optimistic simply because it's software you build once, and then it kind of just gets sold multiple, multiple times. That's the AI model too. You build once and you get so many times.

And I must say the other thing that I've been thinking about, which is on the market itself, not on their competitor, but more on who are they trying to replace from a market perspective. So if you compare technology budget to, let's say, headcount budget, if you look at a normal large company, normally their number one cost is headcount. Normally.

And if you have headcount as number one cost, we're saying that there are a lot of AI agents that can make things a lot more efficient, which means headcount cost can come down. So a lot of costs can be reduced using AI that you build once and then it just keeps running.

So from a market perspective, if we believe in the idea of AGI, if we believe in the idea of agents, it's actually quite big, bigger than we would be able to think today. So those are some of the arguments on why I thought that the 150 billion made sound a little bit mad, but then it's also, there's some hypothesis at the back of it.

Market Entry Challenges in Healthcare

Rod: Yes, you mentioned automation and replacing part of the headcount with agents, all our type of AI processes and so on. And every time we talk about Klarna, there is some news, there is some development in that area where they have achieved efficiencies by removing humans and then instead implementing some type of process.

But in the end, resources are limited, right? So even if you have the possibility to do multiple products, then you have limitations. You do not have an unlimited number of developers. Also, your attention as a leader cannot be on 50 different products, but rather needs to be on a few and so on.

So, Chris, when you're working with teams and trying to decide what should be the features or the products that the company should be offering, how do you balance that in saying, hey, we need to have more to cover more areas of the market. It needs to be a platform, a full-fledged suite. While at the same time saying, hey, we're a startup and we do not have 50,000 employees, but maybe we have a few dozens. And then we have to find out how we can provide an added value product within these limitations.

Chris: Honestly speaking, Rod, I think this is a very natural evolution. As Max pointed out, I don't think that any successful company will only rely on one single product, right? Like it's gonna diversify, I think. And then in the case of OpenAI, obviously I think there's a very interesting situation here.

I think one of the articles was also mentioning that they basically make money through this $20 subscription and they're enough people - I think 10 million or so - that are using ChatGPT on a daily or monthly basis. And that's how they make money. But I think, and obviously we don't know because we don't know the kind of contractual agreements they have with enterprises.

But I do believe that enterprises are still probably a really huge opportunity, probably also largely untapped for OpenAI. But that to me is obviously a second business unit and you would have to like, you know, hire professionals that know how to deal with B2B sales, etc., to make that happen and to also see for whom you build more customized solutions than what you build for the mass market.

So to me, that's really like one area that they can go in, and especially once they're not a nonprofit, but actually like a for-profit company and are professionalizing.

I think coming back to the earlier point on people and organization, etc., and sort of the curve or the maturity curve of a company. In my mind, I actually think that all these really visionary entrepreneurs and founders that call themselves CEO at the beginning of the company most likely should actually move into a chief product officer kind of type of role, right?

Because, you know, think about it. You have Google, you have the two founders from Stanford kind of like tinkering and then building Google out of a sudden. And then I think Eric Schmidt took over the CEO so that Larry and Sergey could also like work on other products.

And I think that that is probably the biggest challenge in my opinion, that normally these people also don't like to deal with politics and people and all that. They really love to build products. And I think that's really what the struggle is once you grow as a company and all of a sudden you're this entrepreneur that needs to be a CEO, and then they don't know or they don't want to, right?

I think another alternative, and I think especially in the case of OpenAI where there's so much brand value tacked along Sam Altman - I mean, right? Like when he was ousted, I think the valuation of what, 86 billion like dropped to zero and everyone was in panic mode because there's so much brand value tacked on Sam personally.

So leave him as a CEO, but make sure that you find a really good COO to really take care of all this internal stuff, all the processes, all the people, make them happy, build a culture. All of that, I think, would be, in my opinion, something to really focus on.

Product Readiness and Release Strategies

Rod: It's always the idea of bringing the adults to the room and so having someone who can be like a steward while, as you're saying, the founders can focus on innovation and so on. But not only do we have the challenge of many products in the company, but also product readiness.

According to this article from Fortune, one of the reasons why the CTO of OpenAI, Mia Moradi, might have left the company was due to pressure on releasing products that many in the team felt were not ready for prime time. So there are some reports that, for example, when GPT-4, one of the latest model releases from OpenAI came out, they only had some days to really test it for any type of malicious use cases, for any type of abuse and so on. And the team felt this was not ready, and yet it came with the pressure from the top of saying we have to go live.

But on the other side, if we recall throughout the history of the show, we have had so many announcements from OpenAI. Over the last month, there was this marketplace where people could make money by providing their own little chatbots or little AI applications. They also had these very impressive demos on generative video where you could create almost like complete little movies just from some prompts. But none of that, we are seeing it right now. And it will be soon almost a year since some of these announcements were made. And it seems the products are not yet ready.

On one side, of course, you do this because you need to keep the market excited. You need to keep investors like, here still dreaming about how the future will be, but also on the other side, you need to have maturity in the product.

And Max, when do you feel is the right moment to launch a product? When should you make the announcement? Should it be something that is still half-baked or you make an announcement when it's 90% ready? Or maybe just try to create some hype and then hope that people will forget and will not hold you accountable that you did not release?

Max: I think that question is a - I think I need to contextualize that a little bit because there are some markets that you cannot just launch and then just forget about it. For example, working with enterprises, for example, you have like one chance to pitch to the CIO or some of the larger CTO companies. And then you figure out what will happen next.

I think, especially when you're trying to sell into larger enterprises, which when you're OpenAI, you're trying to work with like Microsoft to get there, your product needs to be good enough. You have like one chance if you're really, really small. But then if you're OpenAI, if you're that big, they know you have all the resources to throw behind it. Some of them are more willing to work with you to create the product itself.

So that's why I mentioned that it needs to be contextualized in the sense that understanding your own, I would say, positioning in the market and how you are compared to where the market is. So today, OpenAI to a certain extent, they're still in a league of their own, right? They're not really - they have close competitors, but they're not a lot of competitors that are reaching the other level yet, because for them to get there, they will have to probably burn as much money as OpenAI, if not more. And everyone is fighting for GPUs.

Like you said, there's limited money, there's limited resources to get there. So to answer your question then, when should you launch one product really depends on who you're targeting. That's how I would see it. I normally like to see, I would say, let's say B2C products you launch as fast as you can, and then you just iterate really quickly. Whereas anything that has one chance to make it work, else you have to wait for another three years to come back. Those you want to be, just a little bit cautious, right? Just launch just enough so that you could get to the right side of the adoption curve, if you will. That's how I would think about it.

Healthcare AI and Market Entry Strategies

Rod: Chris, let's hear your thoughts as well. And actually, so my question for Chris will be, will you invest in OpenAI? Yes or no and why?

Chris: Yeah, sure. I mean, it's the question of the right price always, no? But actually, I think there are these secondary market platforms out there. And I think there was one where they tested the appetite of the users whether or not they want to like buy a share in OpenAI. I don't know exactly. I didn't look too closely on the price and sort of like which route, etc. But I think there was quite a bit of interest and a wait list because the rounds are oversubscribed.

In general, yes, I think it would be quite nice to have a share in OpenAI at the right price. And I'm not sure if it's too late for that, to be honest, if you can even get your foot into the door. And the reason is simply, I think there are two worlds here. Either there's this like doom world and everything will crash because, I mean, we talked about NVIDIA before. There are not enough use cases or there will be some kind of regulatory restriction coming out. I mean, Europe just announced their EU AI Act, right? So maybe these regulatory trends will actually catch up to the development that we're seeing on the technology front.

So that's sort of like the doom scenario that maybe some of the people are actually subscribed to. But then there's also this very exciting future around, OK, well, we see AI agents. This is just the beginning. The way it grew over the last 1 and 1.5 years is quite exciting. Maybe it's not going to grow with the same pace, but still there's so much room, so much potential for improvement.

I think the technology is still lacking a little bit of consistency on reliability. We talked about hallucination, etc. So because of that, because there's this potential for future growth, and because I think there's a lot of attention in terms of funding, this is the reason why I think there's still, we're not at the end of it yet.

Max: I think just to follow up on the doom scenario like I'm trying to think, what would go wrong? Like when will that 150 billion just evaporate? Right. Will it be No. 1, obviously, like I mentioned, you know, just the free open source model, especially on the Meta side, kind of just upend the business model entirely. No. 2, will be, you know, I would say people like Google, Salesforce just forced their own AI model. No. 3, will be like, I don't know, ChatGPT-5 when it comes out is pointless. Nobody wants to use it. It's very bad.

And then the question becomes, is ChatGPT at the moment good enough for us to do what we are trying to do? And I guess those are the questions I'm trying to run in my head from an investment standpoint to Chris's point. In the right price, will you invest? And what's the downside of investing?

Chris: I think I'm going to quote like Bill Gurley or something. Was it Bill Gurley? I can't remember who. So one of the top VCs said this, right? You can only lose one X of your money if you're going in really, really early. So the right price, the right price.

Rod: And with that, we can move to our next segment, where also is about the right price, where we're talking about generative AI for healthcare. There is a company, so they caught my attention when I saw that, its name is Hippocratic AI. It is a company that provides agents, AI agents for healthcare. So where you have specialized medical healthcare experts that are not actual humans, of course, they are agents, knowledgeable on a specific type of specializations, let's say like menopause experts or assisted living experts and so on and so forth.

And the company is only in its series A and already valued high at 500 million. Max, can you tell us a bit more about Hippocratic AI?

Max: Yeah, so from my understanding of Hippocratic AI, they are trying to build, I would say, almost like a virtual assistant that doesn't require any sort of - doesn't do any sort of diagnostic activities. So that is quite interesting to me simply because from a processing slash operational standpoint, before you even get to the right doctors, there are a lot of administrative tasks that need to happen.

So it's quite interesting in the sense that the angle that they have taken, because for you to be that person to triaging to almost like your front desk help before you get to any sort of doctors or nurses to look at you or healthcare professional, they need to understand a little bit about your symptoms. They need to be able to talk to you in a relatively calm manner, making sure that they are routing you to the right specialist to talk to.

And that also is the job, unfortunately, that less and less people take up nowadays simply because from a healthcare perspective, healthcare organization perspective, they don't see that as valuable or as valuable compared to the actual treatment itself. So the AI, I would say, play around this is quite interesting. I would almost call it like, if you were to call a call center for customer service or help, this feels like it, Hippocratic AI for me is that, helping you to get to the front desk and speak to the right people before they route you to the right treatment or whomever.

Rod: Absolutely. So also along these lines, Bessemer Ventures released recently a roadmap on healthcare AI where they estimate that the spend on hospital care is set at 1.4 trillion, right? We're talking here about massive amounts.

But Chris, I know that you have been exploring the life sciences space and you're aware of the challenges behind. Can you walk us through how easy or difficult it is to establish products or ventures in the healthcare space?

Chris: It's to me, it's a two-sided coin, I think. I was always very much attracted to the healthcare space, and I think this is really coming from the idea of improving lives and saving lives, and that AI could do so much good in there by analyzing data. I think the application of AI and pattern recognition in genetics, for example, could save so many lives, could help on the research front when it comes to cancer or whatnot.

And I think this is very admirable and really something where I think big data, AI, Gen AI now actually really have a right to play. I think on the other side though, it's one of the most regulated industries that exists on planet Earth. And I think that makes it very, very, very slow and also at times probably also very frustrating.

I think you're also just depicting this very complicated landscape of stakeholders you need to deal with and also the many regulations. So when you think from a venture point of view, you know, really building a business that gets adopted by the industry and that you can actually scale across different country borders is really the challenge here.

And I really salute to all the founders that are in the space and that actually do have the patience and the grit and the persistence to really try to move the needle just a little bit in the industry.

Rod: Yes, exactly, so it's a space where always when I go to hospital, I'm surprised how so many processes look archaic almost, and you still see all the equipment and so on, but still, in most countries the situation where healthcare is not exactly affordable or cheap.

And if we look, for example, also at McKinsey, they had also a report where one of the things that caught my attention was just the all the use cases that are available there. So you see that on one side, yes, it is hard. Yes, you need to require a lot of grit, but at the same time, there are so many opportunities.

And Max, as an investor, I'm sure that you keep seeing pitch decks and founders who want to do things in this intersection of AI and healthcare, science, and so on. Are there any segments or sectors that you consider specifically exciting? Where you say maybe the threshold of entry might be lower or maybe here there is a good defensibility but at the same time it's not impossible to access this segment.

Max: Yeah, do I see lower barriers to entry opportunities? They are, but if it's lower barriers to entry, which also means lower barriers to entry to other competitors. So probably not a super good idea if you're not fast enough or you're not ahead to build in those spaces.

I think within healthcare, if I were to think about the entire value chain perspective, a lot of healthcare providers today are slept with, as you put it, a lot of it comes down to cost, ability to integrate from one end to another, streamlined data access. I felt that the biggest value that we could do is actually using some sort of AI agent to be able to normalize some of the underlying data rackets. I think by unlocking some sort of play there could really help, I would say, the delivery itself as well as the overall customer experience.

Max: I can give you an example, right? I broke my leg, I don't know, seven, eight years ago. I went to the NHS trying to figure out what was going on with my leg. I didn't get an appointment until six, eight months later. And then I went and got an x-ray with a private medical, but it's tied to NHS data. But then the reason why I need to do the x-ray is because I then need to go and see the specialist for my leg.

So for like six, eight weeks, my leg is limping. But then I have to take the x-ray, go to the hospital. When I sit down with the doctor, the doctor said, "Oh your x-ray hasn't loaded yet. I can't see it. I can't do anything about it. Come back in two weeks."

Little things like this kind of just drive me mad. Thank God, you know, I just tore my MCL, but imagine if I have a broken leg, imagine someone with way crazier symptoms, what are you going to do with that? I think those kinds of moments, I feel like technology could solve and AI could help a lot.

Everything from scheduling to normalizing data, to sharing of the data, to making sure, providing some sort of, I don't know, diagnostic and support to the doctor. I think that entire process could be a little bit better. So it's just I think taking it from my own personal experience.

Sorry, just a quick one on the contractual. I think one of the big, big problems with NHS from my understanding is the way you would contract software to NHS is very different because they run in like rings of healthcare, like a decentralized centralized manner, if you will. So when you contract with them, it becomes a bit of a problem because the funding comes centrally, but then the need comes decentralized. There's a whole host of problems, but that's more of the healthcare operating model in the UK rather than technology.

Rod: This topic of market entry was exactly what I wanted to address. And I was wondering if Chris has some thoughts about that. When I see, for example, founders who say, "Hey, I have developed this app for doctors. It will make their life easier and so on." But very often, all these stakeholders, hospitals, doctors, healthcare providers, well, they're not exactly at the forefront of technology. They're not necessarily techies, to call it in some way.

And also if I look at hospitals, they might not have necessarily an IT department. Their buying processes might be focused on purchase of medicines and other type of daily supplies rather than technology.

So, Chris, how do you structure the market entry for a product where you have, where you're selling technology, but the stakeholders might not be necessarily technical, and also where you have all these bureaucracies and very integrated buying processes?

Chris: Honestly speaking, I think my previous answer on AI doing good or it's very hard to integrate it into the industry was actually described from the perspective of a startup, I would say, because I actually do believe that AI can do so much in the industry.

And I think it really depends on the perspective because I think in research, AI has been deployed and it's been showing quite some value in discovering new proteins, discovering new chemicals, synthesis and whatnot. And this is something that probably mostly benefits researchers at universities, but then also bigger pharmaceuticals that actually can speed up their product pipeline and research and R&D. I think that's really one way or one value that I see with Gen AI.

When it comes to healthcare and innovation plus Gen AI, I think it's the bigger companies and the incumbents that will have an advantage just simply because they are already in the industry. They have the established distribution and also I think they really have the relationship within the industry. What they lack obviously are these maybe let's say intrapreneurs, but also here I believe lots of companies are starting to build up these capabilities, whether or not it's a VC or sort of accelerator type of program.

I think when it comes to startups, one entry point is definitely this whole consumer wellness space. I think that has been around for quite a bit of time where, you know, maybe you have some kind of wearable and you kind of interpret data, etc., or you try to like actually build software that's actually other industries, very standardized, like a scheduling software for nurses to come into the hospital or something like that. That's not like rocket science, but it's because the industry is so traditional.

That's why even cloud-based solutions with a bit of AI could already do much. Here, I think, again, as I described for startups, it's a big grind. It's long sales cycles. It's a lot of grit and patience and hopefully also very trustworthy investors that are in for the long term.

Rod: Yes, so the idea of maybe partnering, doing maybe like a joint venture with an established player is something that can be a good way to enter the market.

And I would say with this, we can wrap up this episode where on one side we had a long discussion on what's going on with OpenAI, departure, product launches, product portfolio. Should we invest in OpenAI? Yes or no? The challenges that high-growth companies have and how they can master them if they're in a situation similar to OpenAI where culture is shifting dramatically, but at the same time, you still need to keep investors happy and raise more money because, well, you also have a very high burn. You are also spending a lot of money on new development, innovation, and so on.

And we looked into Hippocratic AI, this company that is doing agents. It is already valued at 500 million, although it's very young, Series A, and what it means to launch products in the healthcare and life science space. Not an easy segment, but definitely a space that has potential in really making a change in everyone's life.

So with that said, thank you everyone for being here today. Remember to like, subscribe, give your comments, follow us on social media, and join every Monday for a new episode of the Chris Rod Max show. See you next time.

Chris and Max: See you Monday!

E21: Is OpenAI's $157 Billion Valuation Justified? Chris Rod Max Weigh In