In this episode of the MEDITECH Podcast, host Rachel Wilkes engages in a thought-provoking discussion with futurist Zack Kass, exploring the profound impact of artificial intelligence on healthcare and beyond. Zack, drawing from his extensive experience as Former Head of GTM at OpenAI, shares insights into why AI is personally important to him and recounts transformative moments where AI influenced his life, emphasizing the need for embracing technological advancements. He paints a visionary picture of a future world empowered by AI, offering key insights into ethical considerations and bias mitigation in AI development. Zack stresses the critical relevance of AI for healthcare workers and leaders alike, urging clinicians and executives to recognize its potential in enhancing patient care, operational efficiency, and strategic decision-making. He highlights the necessity for ongoing education and adaptation in the AI field, recommending resources for staying informed. Looking ahead, Zack predicts AI's evolution over the next decade, envisioning its profound implications for healthcare, businesses, and society, and emphasizes the necessity for future-proofing skills in the era of AI.
Zack Kass is an AI Futurist, keynote speaker, consultant, and the former Head of Go To Market for OpenAI. He has spent over 14 years in the field of AI with companies like Figure Eight, Lilt, and OpenAI. His mission is to ensure businesses and governments are active participants in the AI-powered future by demystifying AI, making it accessible and understandable for everyone, and helping leaders navigate the rapidly evolving environment.
While Head of Go To Market at OpenAI, Zack worked at the helm of the team key channeling OpenAI’s innovative research into tangible business solutions. Zack built the teams responsible for OpenAI’s sales, partnerships, and customer success and served as a personal advisor to countless executives deploying the technology across their business.
Today, he is considered one of the foremost thinkers in Applied AI. His thought leadership and techno-optimist views have been featured in countless publications, such as Fortune, Newsweek, Entrepreneur, AdAge, and Business Insider to name a few. Zack has also advised many top companies on AI, including Coca-Cola, Morgan Stanley, and Agmen.
Beyond the corporate world, Zack is a staunch techno-optimist, an advocate for AI’s potential to revolutionize the human condition and reduce suffering. He works tirelessly to promote the promise of AI, serving as a counter-narrative to the prevailing media cynicism.
Zack received his BA from The University of California, Berkeley, and resides in Santa Barbara, California.
Rachel Wilkes is the Director of Corporate Brand and Lead Generation Marketing at MEDITECH. She oversees MEDITECH's corporate communications, branding, events and market research, and serves as executive sponsor of MEDITECH’s generative AI efforts. Over the course of her MEDITECH career, she has also served in various roles in Marketing and strategic Product Management. Rachel is a Certified Product Manager and holds an MBA from Bryant University.
00:00:0:00 - 00:00:07:15
we just need to start challenging the technology companies and the administrators to adopt this,
00:00:07:15 - 00:00:20:12
quickly to actually improve everyone's outcomes. And starting with the doctors and the nurses and say, hey, let's make this job much better. We'll inevitably lead to everyone's improved,
00:00:20:12 - 00:00:21:10
outcomes.
00:00:21:12 - 00:00:30:10
Welcome to another episode of the Medi Tech podcast, where the leader in health care technology empowering you to be a more informed health care consumer and provider.
00:00:30:11 - 00:00:51:05
Hear the latest from our friends and colleagues on topics we think you should know about. I'm Rachel Wilkes, director at Medi Tech, and today I'll be your host. I'm joined by AI futurist Zach Kass. Zach has spent 14 years in AI, most recently as the head of Go to Market for open AI, where he built ad sales partnerships and solutions teams.
00:00:51:07 - 00:01:02:02
Today, his mission is to champion a future of abundance by demystifying AI and helping leaders around the world prepare for the impending technological revolution. Welcome, Zach.
00:01:02:02 - 00:01:04:00
Thanks for having me.
00:01:04:00 - 00:01:08:08
so you bring a wealth of expertise to the field, and I'm curious how all this came to be.
00:01:08:10 - 00:01:33:21
So could you tell me why is AI so important to you personally? And can you share a time where AI significantly impacted an outcome or event in your life, and what your takeaways were from that experience? Yeah, I would say broadly, you know, I got into this field a bit accidentally. I got a job at a data labeling company out of college and, never looked back.
00:01:33:23 - 00:02:05:12
And circa, you know, 2017, it became very clear that I had picked, you know, again, accidentally, probably the most important industry that I could have stumbled into. and then, you know, eventually found my way to to OpenAI. I think that what we are observing right now is the culmination of of a bunch of advancements in the industry, namely the, the the discovery of the modern transform or most of the reason that anyone cares about AI right now is because we started rebuilding.
00:02:05:14 - 00:02:25:06
We we reimagine the way that we design these models. And the modern transformer, which was discovered in 2017 by a Google researchers, is is the most seminal moment in this. Of course, the explosion in the amount of available data and compute, which sort of grew by an order of magnitude over the last ten years, was critical to all this.
00:02:25:08 - 00:02:39:06
AI has changed my life, not just because I happened to work in the field, and it's catapulted my career and all sorts of wonderful, abundant ways. but it also has materially improved a number of outcomes.
00:02:39:06 - 00:02:47:04
Unknown
Let's be clear. Most of what we do day to day is now augmented by AI. We just we just don't realize it.
00:02:47:06 - 00:02:55:03
The cost of goods and services have declined. you know, around the world, because we are better at doing things like,
00:02:55:03 - 00:02:57:20
Unknown
load management and
00:02:57:20 - 00:03:14:12
demand forecasting. Right. We survived the pandemics supply constraints in large part due to, like, incredible advancements in shipping logistics that, you know, helped burden the weight of this, of this really old infrastructure.
00:03:14:14 - 00:03:35:20
and help, you know, nations like the US survive what could have been catastrophic outcomes day to day. Yeah. And it's and so I say all this to remind everyone that we are all massive beneficiaries of advancements in machine learning and and computer science and AI. We just don't realize it because so much of it happens passively, right?
00:03:35:23 - 00:03:57:21
We were the were the second hand beneficiaries. But, you know, today and by the way, we should talk about this. So much is happening right now in research and development. So, so in the medical field, we are seeing huge indicators blinking green. You know, we discovered our first antibiotic in 60 years three months ago because of AI.
00:03:58:03 - 00:04:25:03
We split HIV out of DNA three weeks ago. because of AI. So we are all we are all the beneficiaries. Again of this technology. But I can confidently say today as a solopreneur, right? So I run my own business doing a number of different workstreams. So much of my work now is streamlined because of this technology. I was actually but the response to 80% of my email, I got about 250 emails a day.
00:04:25:05 - 00:04:46:09
And I'm able to respond to a lot of people really thoughtfully. I end up doing an incredible amount of, really intense creative thinking paired with AI. And quite honestly, I haven't used an attorney in a long time. because ChatGPT reviews most of my contracts. Now it's not it's not legal advice to a lot of people.
00:04:46:09 - 00:05:09:19
And certainly if you're working in a hospital right now, you're the general counsel for a very good reason. But, I have the luxury of not having compliance and regulators breathing down my neck and, you know, attorneys are expensive and historically pretty slow. And ChatGPT is pretty thorough and and cheap. So, you know, these are the ways that it that it's changed outcomes for me in the short term.
00:05:09:21 - 00:05:28:02
That's great. That's interesting to you. And you think about the augmentation of creative thinking and what, what I can do there. So, so tell me, what does a world that embraces I look like to you? And, you know, as a self-proclaimed AI futurist, do you have a philosophy about AI that kind of governs your your worldview and the use of the technology?
00:05:28:04 - 00:05:53:14
Well, that's a good question. Yeah. There are, I have a I have a general philosophy that governs my world, really? That has nothing to do with AI, which is that I, I know and we empirically can all observe that the world gets better all the time. so he I studied history and computer science at Berkeley. And one of the things that you learn, if you study even a little bit of history, it's the things that you used to be a lot worse.
00:05:53:14 - 00:06:12:16
And before that they were much worse, and before that they were much worse. And, because today is the best day ever to be born, and because tomorrow will be the best day ever to be born, and because we don't think that that trend will just continue. The real question now is, will we start to see a material acceleration in the improvement of our lives?
00:06:12:22 - 00:06:33:11
And I think the answer is yes. And I ultimately think that AI becomes that accelerant that we are about to democratize and commoditize a number of things that we consider luxuries today. One of those, by the way, is going to be health care. We are approaching a world of abundant health care. We just don't realize it yet. today, health care is a luxury by any account.
00:06:33:13 - 00:06:58:21
certainly good health care. and so, you know what the my philosophy is governed by this empirical understanding that that the world gets better all the time. And the real question now is will it get, you know, logarithmically better? Sure. Fair. And, you know, I know you've talked a little bit about some of the advances in health care that you're seeing and expecting with, with AI, but many of our listeners are health care workers.
00:06:58:21 - 00:07:22:08
And of course, all of us are patients. So what's your take on what's happening around us with regard to health and health care? That requires a creative adaptation of of using AI. And then also, conversely, what happens if we do nothing? What happens if we don't adopt? Well, I'll start by providing the context that I you know, I'm not I'm the child of, to care providers.
00:07:22:08 - 00:07:49:19
My my dad's oncologist, my mom's ObGyn, my, my grandfather was, my, my, my dad's dad was a urologist. My dad's, my, my mom's dad was a was a radiologist. But caregiving has sort of been part of something I've, I've observed for for a long time. And what's really troubling is while patient outcomes are getting better.
00:07:49:21 - 00:08:25:08
patient satisfaction is somehow declining. Patient care is, getting much more expensive despite improved technologies and caregivers satisfaction. Job satisfaction is plummeting, right? This is observable. We have a negative replacement rate for doctors. Why? Because doctors don't tell their kids to become doctors, right? In fact, they discourage it. We have a negative replacement rate for nurses. for all sorts of reasons that we can observe and, and sort of know how if you if you walk into a hospital, you realize how hard this job is.
00:08:25:10 - 00:08:52:16
And so my point, my, my starting place right now is it's untenable. We reached an untenable position where it is so hard to find a happy constituent here. Really accept the insurers and we should. You know, I'll skirt this one very carefully, but if you're a patient, your experience is getting worse. If you're an administrator, your experience is getting worse.
00:08:52:18 - 00:09:10:14
Right. And by the way, if you're a small hospital, bless your soul. You're doing everything you can to not get gobbled up by a big one. And if you're a big one, you're doing everything you can to not get sued for an up anomaly. And if you're a doctor, you know you're doing your best to stay sane. And if you're a nurse, you're probably thinking about changing your career.
00:09:10:14 - 00:09:34:12
And this is the sad state of health care today, at least in the United States. And so what we need is a great leap. Like we just we have to have a great leap. And by the way, I compare health care and education and environment in pretty similar. Like they seem orthogonal. They seem sort of totally unrelated, but they are actually, very interesting commerce situations that are truly untenable.
00:09:34:13 - 00:09:58:21
There is no incremental path to, to a better world. We actually need, you know, great leaps forward and a great leap in this case probably comes in the form of technological improvements, infrastructure, infrastructure change, systematic, systematic change. And we've talked about this for a long time. Right. There's you know, we've talked about health care needing some massive reboot.
00:09:58:23 - 00:10:30:02
The reality is it's too big to have some some policy reboot. So what it really needs is a technological one, and I think represents that opportunity to to massively scale R&D, to massively improve, doctor and nurse efficiency and proficiency and to lift the, the, the base of the baseline of care such that everyone can start to talk about an improved not not not, not worsened experience.
00:10:30:04 - 00:10:54:06
and so my answer is if we do nothing, then we really will run off some proverbial cliff. Right? We will wake up one day with an aged population, you know, say we have a million doctors in the United States. We're projecting to have 800,000 in 10 years. if we don't shoot 800,000, I think in 15 years, if nothing changes, and it's not like we're minting more med schools.
00:10:54:08 - 00:11:18:20
So something's got to give here. and so my message to the industry is, if we do nothing, then we are really. We are truly facing a crisis. and if we think we're facing one today, just wait. Right. So this is a chance, I think, for the industry to accelerate out of, a mess. That, by the way, is really not a fault of anyone's except the weight of the industry.
00:11:18:22 - 00:11:38:02
So I agree. And, you know, certainly we have this tremendously overburdened workforce. I agree that we need to take a great leap forward. But when you're in particular speaking with physicians or other frontline clinician clinicians, what are you telling them about why they should care about AI? So you just gave also really compelling picture of what the industry looks like, what some of the needs are.
00:11:38:02 - 00:11:58:00
But how are you having that conversation with them about why they should care about this particular technology advancement? I will start by saying, if I have to tell someone why they should care about AI, it's not me. It's hard for me to tell someone why they should have to care about AI if they can't extrapolate how far we've come and how far we might go.
00:11:58:02 - 00:12:19:14
I'll start by saying that now. The truth is, most people have come to realize how quickly this is accelerating and people, especially in health care, are smart enough to extrapolate where it where it might go. The real issue is that there is so much cynicism in the industry predicated on how slowly things have moved not outside of health care, but inside of health care.
00:12:19:16 - 00:12:43:03
And so the real issue that I have when I talk to health care providers and administrators is not, hey, hey, it's going to be amazing. It's hey, is I actually going to ever find its way into health care? This is this is the question that that most people have. This is due to, you know, policy and compliance. This is due to to this the speed at which the industry moves all of a bunch of factors.
00:12:43:05 - 00:12:55:04
That being said, my assessment on all this is that there is an incredible amount of cruft computational cruft and bureaucracy that exists in health care that we can't actually throw out. Right.
00:12:55:04 - 00:12:57:02
is a necessity to,
00:12:57:02 - 00:13:08:09
sort of diligent care and compliance. Right. And say what you want about HIPAA. I have some issues with with what it does to actually the burden on R&D, but it exists and it's not going anywhere.
00:13:08:10 - 00:13:38:14
That being said, my dad doesn't need to spend six hours dictating charts anymore. He does it because we haven't distributed the technology evenly yet. But the reality is we have HIPAA compliant ways of recording patient doctor conversations, transcribing them, note taking, setting follow up, distributing to primary care physicians and nurses, and aggregating the notes for for the patient file.
00:13:38:16 - 00:14:06:00
That alone changes doctor and nurse satisfaction. And that technology is exists today, right? Today we can record these conversations in 180 languages and provide thorough notes. If you extrapolate a little on where this is going, people ought to start caring on an R&D basis. I mean, my my thesis is we will cure cancer and neurodegenerative disease within the next 20 years.
00:14:06:02 - 00:14:10:21
So there is there is a whole lot of juice to squeeze here.
00:14:10:21 - 00:14:18:12
we just need to start challenging the technology companies and the administrators to adopt this,
00:14:18:12 - 00:14:31:10
quickly to actually improve everyone's outcomes. And starting with the doctors and the nurses and say, hey, let's make this job much better. We'll inevitably lead to everyone's improved,
00:14:31:10 - 00:14:33:10
outcomes.
00:14:33:13 - 00:14:49:13
Yeah, I agree. And I think what I've seen from my perspective working here at Meditech is this degree of cautious optimism. Certainly understanding the the upward capability of this technology to reduce burden, but also having a fair degree of, okay, well, what are what are some of the other implications? One of the things we should be thinking about.
00:14:49:15 - 00:15:14:08
And I think one of those are the ethical considerations. So from, from your perspective, for the health IT community, for for vendors, for even for organizations to, deploying health care I.T solutions. How do we ensure that ethical considerations and biases are addressed in both AI development process, developing the technology and in the deployment at individual healthcare delivery systems?
00:15:14:10 - 00:15:58:15
Well, I have a hot take, which is I don't think we need any new policy. Hospitals, medi meditech and technology providers are sufficiently governed today by laws that are pretty comprehensive. With respect to ethics, I mean, morality, do we? We are very careful about what we allow companies that provide health care in the United States, and those who support health care providers to do ethics is a is a popular button to press with respect to AI because, there are lots of concerns around, new ways to break new laws.
00:15:58:15 - 00:16:34:10
And my point here is simply we have made it very clear what doctors can and cannot do, what nurses can and cannot do. An AI doesn't actually introduce new issues. What it does is create new complexities in how we govern those laws. And so what's really critical here is making sure that the technologies that are being utilized are HIPAA compliant, are sufficiently secure and private that the patients are aware that these technologies are being used.
00:16:34:12 - 00:16:58:08
There is no world in which a patient is going to be diagnosed anytime soon by AI. That's not something that a doctor is going to do. And it's also, by the way, not something that is legal. I mean, we have actually already, adjudicated this. So what we really need to do is decide that, in fact, the risks that we're discussing here aren't new, right?
00:16:58:10 - 00:17:30:10
They are just being presented to us again. But we've already basically stipulated what doctors and nurses can and cannot do and that we should, you know, effectively, protect for these things. And by the way, what isn't ethical is often also just bad practice. So, for example, we may not actually have ever considered whether a doctor can, spoof their own voice to tell a patient about a prognosis.
00:17:30:10 - 00:17:54:10
That's probably not something that we've ever, adjudicated. It's also not something that a patient would want. And so I think what we also need to be honest about is like, look, let's let's find all the efficiencies in the system and also regard for patient satisfaction and outcomes, which would should be in the interest of the administrator and the and the caregiver.
00:17:54:12 - 00:18:17:22
That's fine. What I think, where there's a lot of hand-wringing and I frankly don't think there should be right now, is around this idea of biases in the system. And what I remind everyone is, if you think AI is biased, go meet your neighbor. The reality is, what's what's unique about AI is this idea of explainability. Humans lack explainability.
00:18:17:22 - 00:18:39:05
And this is such an important concept, right? There is not enough therapy in the world for an entire hospital of doctors to understand their own biases. Right. We are. We are creatures of incredible complication. And this idea of, you know, we throw out around these terms of trauma and triggering these days. But it's true. We grew up with all these events that lead us to believe certain things.
00:18:39:07 - 00:19:05:23
A doctor would never, ever begin to understand why they believe something, right? It would take years and years of incredible psychoanalysis. Moreover, they would never actually admit that thing. They would never say to a patient, hey, you know, I, I struggle in the O.R. sometimes because I had a broken relationship with my father, right? This is these are real things that happen in everyday lives that we can't actually process because it's not appropriate, that we're not capable of it.
00:19:06:01 - 00:19:39:10
AI is capable of actually explaining the decisions that it makes. So if I makes a recommendation or, you know, conclude something, it's actually quite capable of saying, here's why I did this. AI is far less biased. Therefore, then humans and this idea that we are introducing new bias is sort of funny to me, because it's potentially the first system that we've ever explored that has human intellectual equivalents without human intellectual, without human bias equivalents.
00:19:39:12 - 00:19:51:15
And, you know, for the first time ever, we might be able to have a system that, like, doesn't actually consider all of the things that humans consider when it makes really critical, important decisions.
00:19:51:15 - 00:20:06:19
You know, it's an interesting way to put it, to use a human intelligence without the level of human bias. And given that and I don't I don't know how many people are thinking that way, but if you how do you have organizational leaders foster a culture of trust within an organization for AI?
00:20:07:01 - 00:20:24:19
So you've talked about explainability. You've talked about AI, maybe not having the biases that people, think they that humans think. But I kind of guess how do you I guess I would the psychology of getting through that and making sure that people are okay and just getting to the point where they do trust AI, particularly health care.
00:20:24:21 - 00:21:04:05
Well, it's a yeah, I mean, this I don't know that any traditional systems are inherently skeptical of of change, which is a feature, not a bug. and and you know, tech knows that. Well, right. Introducing a new products or a new feature into, into a medi tech software is a very, very, you know, I'm sure, journey with the hospital because you're talking about years and years, decades of, of of behavior that is, you know, hard to change.
00:21:04:05 - 00:21:31:01
And having observed my parents, doctors are not really technologically savvy. And that's been no, no fault of their own. It's just not the skills you learn. And so introducing change is hard period. So how do we best do this. Well in criminal improvements that have material outcome are the easiest way to do this. And one of the things that we should be doing is finding the highest pain and lowest risk points.
00:21:31:03 - 00:21:51:15
And obviously that's in no taking. I mean, this is why I bring this up over and over again. Note taking is this exceptionally high pain and exceptionally low risk, place. Because if you're if you're recording the conversation, you can go back and re listen. Right. Just because the transcript arrives and just because you're 99% sure that's the transcript, you can still re listen to the conversation.
00:21:51:15 - 00:22:22:14
So it's actually introducing redundancy, and safety starting at a place like this allows the system to slowly maneuver and acquire trust in what will eventually be sort of sweeping improvements to, to practices. But, you know, starting at the again, the Venn diagram is where did where do people experience a bunch of pain today? Where are we willing to explore this technology and where's the outcome?
00:22:22:14 - 00:22:48:06
Obviously going to be going to be really good and you know, this is this is how any system adapts to change. But in particular, one that's designed to move very slowly or update very slowly, like a hospital. You know, this this is exactly how we how we should be introducing, this new technology. And it's, I think one of these things where a doctor is going to get this is it's going to change or a nurse is going to change your life, and then everyone's going to go, wait a second, how do I how do I get this thing?
00:22:48:07 - 00:23:09:09
And then the administrators aren't, you know, instead of the ministry just trying to push trust, the administrators are actually going to be feeding, you know, we're going to we're going to flip the script suddenly, and, and, you know, the technology that we introduce is going to be seen as this welcome versus this, you know, rejection. And we've seen quite a lot of that too.
00:23:09:09 - 00:23:27:07
So our own experiences that the note taking documentation has been an area when the first kind of set of use cases that that we've tackled, because the same thing, it's a highly burdensome, intensive area that we feel like once you get some wins there that can help, you know, build, build some trust and, and competency in the technology.
00:23:27:09 - 00:23:49:18
so I know you talked about I shouldn't be doing any diagnosing. Right. But from your perspective, what are the any other things I in health care can't or shouldn't do? Well, I'll start by saying I have no issue with AI doing diagnosing. but I don't, but I, I don't fall in the this is I'm an outlier and all these things and I accept that I'm an outlier.
00:23:49:18 - 00:24:18:13
So I never use myself as a straw man. I the limit. Let me tell you what I believe. At the limit. Health care is fundamentally a game of relationship. It is managing, and it is an individual managing an individual's expectations and optimizing their outcome as best as possible. And at at its best, health care is about a doctor and a patient, a nurse and a patient.
00:24:18:15 - 00:24:52:12
And or we should say broadly, caregiver and patient and that patient both understanding what they need to do, understanding what will be done for them, and that that plan being executed as effectively as possible. Everything that happens outside of that, which unfortunately, is so much of health care today should be automated. Everything finding a doctor, scheduling with the doctor, the the paperwork leading up to and after.
00:24:52:16 - 00:25:14:00
I mean, I am shocked. And by the way, I am reminded all the time of how amazing the health care system is when I can't call in a favor from my parents because so much of health care today is knowing someone and I and I see this now, I mean, with an incredible amount of privilege. You know, it's really upsetting.
00:25:14:03 - 00:25:34:19
It's it's so sad. If I can't call a fever, it takes me a long time to see a specialist. And you know, moving back. I moved back home to Santa Barbara. And one of the incredible luxuries of moving back home to Santa Barbara's, you know, all the people you need to know. And if you don't know someone, I mean, it's hard and that needs to that needs to go away fast.
00:25:34:20 - 00:25:59:13
And all of that stuff should be automated. And all of the follow ups and the, the, the the post ups and the and the and the check ins should be automated and we should get to a place where if a machine can do it consistently and, and, and, you know, it doesn't, it doesn't have a crazy margin for error.
00:25:59:13 - 00:26:28:17
It should be done. And what we should build is a health care system where what humans are doing is the highest leverage work. Having hard, interesting conversations with patients, delivering the most important care and change, doing things like changing lines around patient behavior, delivering really challenging news about prognosis, delivering really good news about prognosis. The things that we want humans to do.
00:26:28:18 - 00:26:47:09
And, you know, I'm reminded all the time, you know, people fly all over the world to see my dad. And oncology at this point is a pretty automated practice, right? You you you you, but you plug in a bunch of diagnostics. It comes out with a, with a recommend, a recommended, medical plan. So why do people want to see my dad?
00:26:47:15 - 00:27:12:16
Well, he's a, an exceptional listener. And it turns out when you're facing a life changing crisis, which you want to someone exceptionally empathetic, his patient outcomes are probably marginally better, in large part because he probably puts patients in a better mind space. But oncology is is is more and more being reduced to empathy. I mean, if you go to school for a long time to simply be a really kind person, who knows?
00:27:12:16 - 00:27:32:16
Happens know a lot about about cancer. That's how a lot of this is going to go. And obviously surgery accepting so much of this is going to be about bedside manner. And I don't think we're going to call it bedside manner anymore. I think we're going to call it health care. I think we are basically going to redefine the importance of being a human in health care.
00:27:32:18 - 00:27:57:01
As long as we can strip out all of the things that aren't actually patient care. And, you know, this is a big vision and which is, you know, I don't lead with this, but, man, it's most of the stuff that caregivers are doing today is not actually caregiving. And if we can if we can, if we can do away with that, we should be left with, you know, the, the highest, the highest value work.
00:27:57:01 - 00:28:02:23
And by the way, why everyone gets into this field in the first place. No one becomes a nurse because they want to fill out charts.
00:28:03:05 - 00:28:27:05
Right. Let's let the passionate people be passionate about taking care of people. Absolutely. So in inside or outside of health care, is there anything about AI that keeps you up at night? Oh, plenty. Look, the the the downside here. And I don't think it's a long term downside, but I think the short and medium term downside is pretty clear.
00:28:27:07 - 00:29:05:04
There are three really weird outcomes. The first outcome is this, theory of idiocracy, which is this idea that, you know, in a world where we solve a bunch of our problems very simply, or I should say very easily, and in a world where we get very good at hyper personalized content delivered at exactly the right time, very, you know, very easily or very inexpensively through, through very abundant machines, there is a very real risk that we simply devolve intellectually because our the importance of critical thinking declines.
00:29:05:04 - 00:29:40:08
And, and the, accessibility and attractiveness of the alternative. You know, let's, let's be frank, social media content but but just sort of hyper personalized content on your, on your device becomes so good that we have this like, you know, k curve in, in the behavior of, humans. And we're starting to see some weird trends in Gen Z that are sort of suggesting, a regression, in socio cognitive behavior.
00:29:40:10 - 00:29:55:05
Now, by the way, Gen Alpha is exhibiting inverse behavior. So I think it it could simply be generation. but we should never giving kids cell phones and we should never have given social media. We should take it all away as fast as we
00:29:55:06 - 00:29:55:16
Unknown
can.
00:29:55:21 - 00:29:56:04
so
00:29:56:04 - 00:30:19:08
that's a clear risk analyst. And that leads to the second risk, which is dehumanization. There is a world in which we actually get much smarter because of AI. we expand our minds. And by the way, I think we do, for what it's worth, like, I think we will expand our minds and our, you know, our children's understanding of the world is so much greater than our understanding of the world and and so on, and that that's been a true trend for forever.
00:30:19:08 - 00:30:45:09
But one in which we actually find more joy in the company of machines than in the company of each other. And this one's weird, right? So whereas the first, the the theory of Idiocracy looks like Wall-E, where we sort of surrender ourselves to machines and, and, you know, find the pinnacle of existence is actually consuming food and content, sort of floating around on beds.
00:30:45:11 - 00:31:10:01
the the second negative outcome looks more like her. Joaquin Phoenix's her right where this man is very lonely. Man finds, companionship and a humanoid robot. Beautiful companionship, by the way. But that obviously has incredible negative consequences. Namely the decline of civilization, because it would, you know, lead to a fertility crisis. But it's also clearly not what machines should be doing.
00:31:10:01 - 00:31:35:17
They should not be providing us compassion and joy. They should be they should be doing all the other things, because humans at the limit should actually just be congregating in community outside, among friends and family. That's probably what we are most meant to do. so those are those are the two risks. And then, you know, the third risk, which is the one that I, I think is inevitable, quite honestly.
00:31:35:22 - 00:32:02:04
Whereas the first two, I think we can avoid with some really thoughtful, management. The third is this idea of a identity displacement crisis. So right now there's all this attention on job displacement, and I think it's misguided. And I don't think it's misguided because jobs will not change and go, come and go. I think it is misguided because the real risk is not the economic implications of your and my work changing.
00:32:02:04 - 00:32:28:20
Right. It will change, that is for sure. what is scary, though, is that. While we may be able to put food on the table, attaching our sense of self and our identity to our work that is constantly moving will be very. And you know, there's a reason the most common last name in the United States is Smith and Miller is because we used to name ourselves after our occupation right.
00:32:28:20 - 00:32:51:20
It would be for a long time in a post-industrial world, we have built an inextricable connection between who we are and what we do. And that has incredible knock on effects, mostly related to GDP growth and innovation. Right? The more obsessed we are about our work, the healthier our economy gets. And it's it's it's a really good thing, but it has incredible consequences for for mental health and emotional health.
00:32:51:22 - 00:33:19:05
And this is why we observe incredible depression among retirees especially. And that trend is increasing, right. Because people believe that they are what they do. And so in a world where you can't pin the tail on the donkey anymore, where your work changes so frequently and so much that you can no longer comfortably say, I'm an accountant, I'm a lawyer, I'm this, I studied for this and I, you know, I'm a company man.
00:33:19:07 - 00:33:53:03
That's going to be really hard. And I think identity displacement is going to be something that we are truly is the thing we are not prepared for. And I think it's going to be the, you know, I think if there if there is and I, there probably will be an incredible backlash to AI at some point. you know, we've observed troglodytes and Luddites historically, people who reject technology or try to destroy technology or don't understand technology, we will have huge movements, industrial movements to sort of stop the spread predicated not on the economic implications, but really on the emotional front.
00:33:53:05 - 00:34:13:01
And I don't know that I, I know we're not prepared for it and I don't know how to play out, but there will certainly be social unrest, that it's all it's all interesting and there's so, so much to think about and so much for all of us to do to keep learning and stay on top of how things are trending and what the new considerations are in this space.
00:34:13:03 - 00:34:33:02
So what are some things like how are you staying updated on the latest advancements? Any trends? interested in any kind of favorite publications that you're keeping tabs on, books that you've read lately, and what are you reading? I don't know who. What what? Yeah, what don't I read? it's a little tricky. I will say it's tricky for two reasons.
00:34:33:04 - 00:35:00:08
There's basically two kinds of content today. there's an exceptionally. There's the white papers that are moving the industry forward, and these come out daily at this point. I mean, truly, we are seeing, research papers for, published on a daily basis that are moving the industry forward across the cross. I yesterday, yesterday, Rachel, we saw two state of the art models published.
00:35:00:10 - 00:35:17:17
I mean, that's how fast this stuff is moving. If you are interested and you are academically inclined, there are a lot of places to read the white papers and to see the latest. You can also download white papers, put them in the chat GPT and say, explain this to me like I'm a 12 year old, which works really well.
00:35:17:17 - 00:35:49:14
And surprisingly, no one's no one's built, an API for that yet. And I honestly, it's like a project that I wonder if I should build. The second thing that you can do is is read the, you know, the zeitgeist in the press, which I think is 90% garbage, but 10% very good. And there are a handful of people that are consistently writing really thoughtful things that sort of cross the chasm, you know, and I there's basically this, a quadrant of thinkers, people who think it will be very good.
00:35:49:14 - 00:36:07:20
People think very, very bad, think people who think it's moving really quickly, people who think it's actually moving slower than we think. And so in that quadrant, you basically have people who, you know, in the upper right where I exist, people think it's moving quickly and it's very good, etc.. there are a lot of places to go.
00:36:07:21 - 00:36:37:17
The three books that I recommend that everyone read are The Coming Wave by Moustafa Suleiman, live 3.0 by Max. Actually two books, I'll Keep It Simple Life 3.0 by Max Tegmark, and The Coming Wave by Mustafa Suleyman. Those are two approachable, good books that give you a comprehensive understanding of the future of a I. I have a book coming out later this year, Meditech can buy one for all its clients.
00:36:37:17 - 00:37:04:11
I'll sign it. and those are written by two moderately optimistic people who recognize all the risks. and, you know, life 3.0 by by Tegmark is really considered, I think that the gold standard in the stuff outside of that, my best advice to people is stop listening to the news. It really it's poisoning minds. It's not helping anyone and they don't know what they're talking about.
00:37:04:11 - 00:37:25:00
Like they really don't. I mean, they're trading on fear. you know, if you turn on, if you turn on CNN tonight, you'd be convinced that the world might end tomorrow. and it's just not it's going to get a whole lot better tomorrow. And the best thing we can do is, is demand that that that mass media start telling us, giving us reasons to be optimistic.
00:37:25:02 - 00:37:45:15
it's just it's the most interesting thing that they could do today. Right. Well, I have to check out those those books. I appreciate the recommendations. They're, again, given how fast everything is evolving here for folks that are looking to either start a career in data analytics or AI or advanced their AI specific skill sets,
00:37:45:15 - 00:38:11:07
for folks in like specifically technical roles, anything that they should be thinking about. Yeah. So if you are, if you're administrator, if you're operating in a very technical role, this is, this is going to sound self-serving on this podcast, please be prepared to adopt the technology that is going to materially change the direction of the hospital and the the caregivers experiences.
00:38:11:09 - 00:38:44:22
We cannot I we cannot live and die by the by the by the excuse that it's a hospital. So it moves slowly. It's just not it's not an acceptable answer anymore. So the buck stops with someone. And so my encouragement to anyone who operates in a, in a highly technical role is adopt this stuff that can that can materially change these outcomes adopted at home so that you can be a you can be a holistic practitioner and adopted at work because people actually want it.
00:38:45:00 - 00:39:12:00
and it will it will materially improve the, the cost and the outcomes of of everyone's care. And please, please do not, hide behind the excuse that hospitals must move slowly. It's it's it's it's a it's a self-fulfilling prophecy that we can that we can break. We have the power to make the systems move quicker. Totally fair.
00:39:12:00 - 00:39:34:18
And my my last question for you, think aligns with, you know, everything that you've, that you've said and how you describe yourself as a futurist. Let's kind of close with how do you see AI evolving in the next 5 to 10 years? And the implications, you see, maybe what's most specifically on health care? Well, we will have AGI within ten years, and that's going to be one of these, you know, seminal moments in the human experience.
00:39:34:18 - 00:39:53:13
It'll that we'll look back on quite like we look back on fire, except that it'll proliferate almost instantaneously. And in that sense, I think we're about to experience the greatest industrial revolution in the human experience. That being said, in the next two years, we will see a gentle boom. The next year we will see a giant of AI.
00:39:53:13 - 00:40:13:22
So, so autonomous agents that will do things on our behalf. And suddenly instead of, you know, imagine a Siri that worked, this sort of what I call people and such that instead of actually opening up our apps and opening up our browsers, we will start to give assignments to this technology. And that's going to be this right brother moment for a lot of people where they go, oh, you know what?
00:40:13:22 - 00:40:34:06
Actually, my life is filled with a bunch of crap and bureaucracy that I don't actually enjoy, and that doesn't accrue value to me. It's just stuff I have to do. And suddenly everyone's going to start to to experience AI in a really profound way. And my theory is whatever happens with this technology in any direction it goes, it will ultimately lead to much more time.
00:40:34:06 - 00:40:56:12
And so my, my, my closing sentiment for most people is start practicing in a very deliberate free time because you're about to get a whole lot more of it. And I'm not convinced yet that we know how to spend our free time. Well, we talked about the theory of Idiocracy and dehumanization earlier. I'm desperate for people to start practicing very, very deliberate free time usage.
00:40:56:14 - 00:41:19:07
limit your screen time as much as possible. Go outdoors, be with friends and family. smile. Do you know, meditate, do the things that you know the healers on Instagram talk about, but actually do them? Because when this industrial revolution arrives, I truly believe we're about to. We're about to see what happens with with a 20 or 30 hour workweek.
00:41:19:09 - 00:41:23:20
and that that's going to have all sorts of funny knock on consequences.
00:41:23:20 - 00:41:42:14
But that's a really and a happy thought to end on and appreciate that. And this was, this is a really fascinating conversation. I appreciate the time you spent with us and all the perspectives you shared. Thanks. So thanks so much for having me. And thanks for all the work that you and all of your listeners are doing to to improve patient outcomes around the, around the country.
00:41:42:16 - 00:41:47:17
Thank you, and we look forward to seeing you at Medi Tech Live at the end of September. Thank you so much.
00:41:52:09 - 00:41:56:04
Thanks for tuning in. Stay informed and subscribe to the Medi Tech podcast and
00:41:56:09 - 00:41:57:18
and we'll talk to you next time.