May 30, 2023: Charles Boicey, Chief Innovation Officer at Clearsense joins Bill for the news. How might the adoption of AI and transformative technologies in healthcare impact administrative costs, job roles, and the overall structure of the industry? What are the benefits and limitations of the consumer-facing approach of the Hippocratic language model in providing information and support for healthcare-related tasks? What are the challenges and limitations associated with scaling up AI models like GPT-4 in terms of size and parameters? What are the legal and privacy concerns surrounding the use of AI models, such as copyright issues, data monetization, and privacy regulations?
"The Future of Care Spaces" is an upcoming webinar that explores the latest healthcare technologies and solutions transforming care spaces in America. Hospitals, clinics, and at-home treatments are all affected by the advancements in healthcare technology, which can improve workflows, treatments, and patient outcomes. What are the latest remote monitoring tools, advanced telehealth solutions, and other innovative technologies transforming care spaces across America? Join us June 8th, 2023 1:00 PM ET and Register Here
This transcription is provided by artificial intelligence. We believe in technology but understand that even the smartest robots can sometimes get speech recognition wrong.
Today on This Week Health.
at some point in time in the future your health will be a hundred percent monitored. You'll be able to see where you are at any particular time.
You won't have to, you won't have to go to this or this. It'll be right out there in front of you. (Intro)
Welcome to Newsday A this week Health Newsroom Show. My name is Bill Russell. I'm a former C I O for a 16 hospital system and creator of this week health, A set of channels dedicated to keeping health IT staff current and engaged. For five years we've been making podcasts that amplify great thinking to propel healthcare forward.
Special thanks to our Newsday show partners and we have a lot of 'em this year, which I am really excited about. Cedar Sinai Accelerator. Clear sense crowd strike. Digital scientists, optimum Healthcare it, pure Storage Shore Test, Tao Site, Lumion and VMware. We appreciate them investing in our mission to develop the next generation of health leaders.
Now onto the show.
(Main) 📍 hey, it's Newsday, and today we are joined by Charles Boicey, co-founder. I, I looked at your LinkedIn and said co-founder, and I didn't realize that was one of your titles. And I think Chief chief nerd, chief Technology Officer, chief Deep Thinking data person over at Clear Sense are any of those titles accurate?
I think that, I think you got my name right, bill
and I do a little bit of this and a little bit of that. Chief miscellaneous
in charge. Yeah. Whatever you yeah. I didn't realize you, you counted co-founder as part of your titles, although I know you are. Because, yeah, we talked about this technology before you actually went over to, to clear sense.
I mean, I know this was stuff that was rolling around in your brain and stuff that that you had that really pioneered at UCI and then took to the next level when you're at Stony Brook. And then obviously then went to clear sense. So you had a lot of time thinking about what it should look like before you built it out at clearsense.
Yep. Yep. Other duties as sign though other duties as a sign. And for people who aren't watching any particular reason you're wearing Hawaiian shirt today, or is it dressed down the newsday today or?
Yeah, since everybody's remote, we had an opportunity to bring the tech team together.
So there's about 35 of us and it's Hawaiian shirt kind of contest
day. So, we'll see who wins. I hope. Do me That is, I was gonna say, that's not the wildest Hawaiian shirt I've ever seen actually. It's a pretty, pretty nice looking Hawaiian shirt.
No, there's a guy here with a Hawaiian shirt that has bananas on it, and he's wearing a matching pair of shorts with bananas.
So I think the banana guy is gonna win this one thing.
Things we miss about not getting together. Well, since you're on the show, we are gonna talk a lot of really fun stuff. We have a glimpse I this rumor has been out there for a while, but we have a little glimpse into Apple's virtual reality headset.
I want to talk about that with you cuz you and I have talked virtual reality a couple times. We're gonna talk about new startup Hippocratic as well. And obviously we've gotta hit generative ai. Because even off the air, you and I have been doing a lot of stuff talking to different companies and talking to each other about generative AI's impact on healthcare.
So, let's start with Hippocratic. You wanna start there?
Yeah, sure. Absolutely. Do you want start? Do you want me to kinda get into it? Oh my
gosh. You would actually do the the lead-in for, I usually do the lead-ins, but I'd love to, I'd love to have you do the lead-in. That would be great. So Charles, the lead in, here we go.
Let's talk about Hippocratic. Everyone taking a look at it get it, get a little bit more involved with it, but it's really important because so Hippocratic has developed a a language model or as we call it, a large language model specifically for healthcare. Why is this important?
It's always been my position bills as well, that these language models have to be curated. They have to be exact, especially for healthcare. If you're gonna combine the all the information, available in the planet. And then, maybe some outside, then you're gonna have some errors and whatnot.
But if we have the proper subject matter experts creating these large language models, we can have one for nursing, we can have one for, physicians, pharmacists, Whatever the sub language model may be, but you know, collectively we have a curated model and we keep it up to date.
Then, that's what we're looking for from a generative AI perspective. That's how, you know we're gonna be helpful and they've done it. So, they got together bill, you might have the particulars on it, but they got together, they produced this and,
They're ready to go.
I do have the particulars on it, so, hashtag a general catalyst. They are, they've raised 50 million. They're coming outta stealth today, and this article was written yesterday. So, Charles, one of the things I appreciated about you is you sending me articles. Usually I send the articles out and we talk about 'em, but you're sending me, actually the first two articles we're gonna talk about were from you.
50 50 million was raised. General Catalyst that's behind them. Andreessen Horowitz. So this is this is big time. They've been working on this for a number of years. So this isn't like, Hey, we raised some money. We're gonna figure this out. They've been working on this for a while.
They have group of physicians, hospital administrators, medical professionals, and AI researchers from organizations including Hopkins, Stanford, Google, Nvidia. After co-founder and c e o Manel Shaw sold his previous company like.com, a shopping comparison site to Google in 2010. He spent the better part of the next decade building out Hippocratic and, oh, has some information down here.
They have some of the missteps, right? You had G P T three recommended. It's a person commits suicide. That's an example of it's been trained on everything, right? So it can give, it can potentially give bad information back out. Babylon Health. Is under scrutiny, the UK's National Health Surface U utilized them and they made claims about disease diagnostics and whatnot that, that haven't come to fruition.
And obviously we all know about IBM Watson Health. So that's the reason for these narrow models. They have to be trained. Very specifically and one of the, one of the cases they give here, let me see if I can blow this up. I can't.
And Bill, as you're looking that up, it goes back to humans can be forgiven.
Machines get one shot at it.
Yeah. Machines don't get to make a mistake. The we get to, yeah we do, we've talked about that with cars, a person, how many accidents will happen today, but if one of those is from a self-driving car, That will be the front page news, whereas all the other accidents will probably, we will never hear about.
They have these interesting things where they take tests right? North American pharmaceutical licensure examination, registered nurse, acute care, certified pediatric
Yeah. Bill, look at the number on the the pass rate on the NCLEX for nursing. I believe it's in the high eighties, correct.
Yes. Yeah. Yeah. I mean, so I mean, yeah, Hippocratic essentially is a b plus a student on all of these tests. To give you an idea, G p T four, which we've been touting for any number of weeks here, is essentially a C student on most of these things. B student on one or two of 'em, but mostly a C student and in some cases they are an F student.
And then you have some of these other LLMs, which are really low. I mean, they're F students.
And you said it before Bill, narrow focused, narrow thing focused large language models.
Yeah. So, yeah. So this model, the Hippocratic model, is not gonna be able to generate your emails for you or write articles about finance.
It's not trained on that. No. Nor really should it be. Well, here's what I found interesting on this though, Charles. Shaw emphasized that Hippocratic isn't focused on diagnosing, so that's interesting. Rather he says the tech will which is consumer facing. So this is a consumer facing tech. Informative. Yeah, informative. So, so it's not doing diagnosing, it's consumer facing and it's aimed at use cases like explaining benefits and billing. Providing dietary advice and medication reminders, answering pre-op questions and onboarding patients and delivering negative test results that indicate nothing's wrong.
So, and there's, yeah, the dietary advice, they go on and talk about some more things there. I think that the point that we want to make over and over again is it matters the data that's used to train it. Right. It, yep. So G P T four, and it's amazing. I use it. It's very it's very helpful.
But I think what we're gonna see is these narrowly trained models. And the other thing I'm saying to people these days is that the think of AI as your minion. And at some point AI is gonna recognize that I'm calling it a minion, but think of it as your digital servant. And I'm looking at G P T four now has plugins and you can plug it into things like Kayak for travel.
And you can say, Hey, what's the cheapest flight for this or that? And it'll just pop it out. Well, it's a lot easier than doing a search on kayak, quite frankly. Doing it through G P T four cuz it just processes all that information and pops it out. It now becomes a digital servant. I can schedule things, I can have it respond, I can have it generate things.
Yesterday Google introduced a feature in Gmail. Did you see this? Yep. Yeah it'll essentially write your emails for you. Absolutely. Like you give it a couple prompts. Absolutely. And it'll pop out the email. So, and
these mass layoffs now will be a little bit more empathetic when it comes from generative ai.
brutal. You're brutal. But, it's interesting cuz that's one of the stories. Let me see if I can find that one. Let's go back to inbox. Oh, and this is Taylor Davis had a post, I'm gonna go here first. So, Taylor Davis had a post, and one of the things he said is his predictions for AI and healthcare next 24 months.
Healthcare lags an uptake of AI compared to other industries, and that's to be expected by 2025. Other whole industries will have been transformed by ai. And there will be setbacks, but in many cases, the companies will shed 30% of their labor costs. Health system CEOs will watch these self-evident and shocking transformations.
They will realize that the race is on whoever masters. The AI transformation first will have a whole different cost equation with which to grow, market, and transform. Expediency will be the norm revenue cycle in particular, we'll see a 50%. Reduction in their current workforce. Few health systems will stumble in dramatic fashion, but public embarrassment is not the greatest risk that goes to the health systems that could not shed 2010 to 20% of their costs, they fall behind their competitors and are acquired.
And this is somebody who's in healthcare, somebody who knows healthcare, who's essentially saying, look, I'm looking at the AI market. I'm seeing what it can do. And I'm looking at healthcare and I think it's gonna be transformative to the tune of at least 10 to 20% of the administrative costs.
And I think it's gonna be the catalyst
for more m and a activity, so we one time we were at 5,800 healthcare organizations, we're now down to, under 4,000 and, eventually we'll be around 200, 300. So, so yeah, those that, don't adopt and, Operationally is where, focus secondary, probably on the financial side, and then thirdly on the clinical side.
So, operationally, hey we need to do something as far as, clinical times of service where they're located. All the operational is where it's gonna, it's
gonna hit first.
So people are worried about the job loss, but you and I are old enough, we have enough gray hair to remember these arguments when the PC came in and people are like, oh my gosh, this is gonna put people out of a job.
And and there's various technologies that have come into play over the years. That people were like, oh, there should be mass layoffs. The reality is the PC came into the world and there was mass layoffs in certain areas, and then there was all new jobs created. Yep. That didn't even exist before, and people just moved into different jobs.
And I assume the same thing's gonna happen here.
Oh, absolutely. And, people are become more, the human side of this thing this technology allows for it's a adjunctive, it's, it's helpful. It's not there to replace, we can do a better job or, you think about call centers.
Do you really wanna operate your call center, twenty four seven with humans, you don't, you're not gonna have to, you don't, you actually don't have to Now,
Well, let me one more. Generative ai then we're gonna talk about Apple's headset. So, okay. Sam Altman, CEO of Open AI dropped a dropped a bomb at a recent M I T event declaring that the era of gigantic AI models like G PT four is coming to an end.
This guy who's in charge of G PT four Anyway, he believes that the future progress in AI needs new ideas, not just bigger models. And he says, look G P T two had 1.5 billion parameters. G P T three has 175 billion parameters. G P T four has closer to trillions of parameters. This kind of parameter growth is no longer tenable.
Feels Altman. Why number one, you said return scaling up models. Size comes with diminishing returns. Two is physical limits. There's a limit to how many and how quickly data centers can be built. And then three is cost Chat GPT t costs over a hundred million to develop. And the person who actually wrote this goes on to say that they don't really think this is the only problem.
And they talk about the access to data. Why is it becoming harder and more expensive to access data? Copyright conundrums, Getty images, individual artists like Sarah Anderson. Kelly McKernan are suing the AI companies over unauthorized use of their content, universal music as Spotify and Apple Music to stop AI companies from accessing their songs or training privacy matters and regulations.
Italy Band Chat, g p t over privacy concerns. Samsung even warned employees not to use AI tools like chat G P T for security reasons, data monetization, Twitter, Reddit, stack Overflow, and others want AI companies to pay off for training. On their data. Contrary to most artist Grimes is allowing anyone to use her voice for AI generated songs for 50% profits, so forth and so on.
So I said a lot of things there. One is, these models are getting into the trillions of parameters. There's We're not he thinks it's not tenable because of the scaling issue. We can't build enough data centers diminishing returns and just the cost. And then this person goes on to talk about access to data, to train these large language models is gonna become more scarce.
I'd love to hear your thoughts on that before we go to the Apple headset.
Oh my, is my comment. Don't you just love chaos? So we're gonna have, we're gonna have a litigation pause probably, but maybe, maybe we won't. Again limited language models, focus language models.
And, that'll bring on the need for interoperability between language model So we'll throw that one out there. But I think these focus language models for specific, even outside of healthcare will be will be a way to, alleviate a lot of this. Also you as we go forward with this, There's advancement in compute, there's also advancement in storage.
So, DNA storage, quantum computing and so forth, that is becoming more and more affordable. Even, as we speak. That will be, that will be help, that'll help all this from a technical side, the legal side, that's all humans. So that's gonna take a little while to to get through.
So you're gonna continue to see that, banning and also the, the heavy litigation.
We'll get back to our show in just a minute. I am excited about our webinars this year. They have been going very well. What I've done is I've gone out and talked to people in the community and said, what works in webinars?
And they came back and said, look, this is what we want. We want a webinar that is not product centric. It's really focused in on the problems of health care. And we want people on there that are actually solving those problems. And so we have done that. And the response has been fantastic this year. We have another webinar coming up.
It is the future of care spaces. Where care is being delivered is changing rapidly. Even the care spaces within the hospital themselves are changing. Technology is being added in different types of technology. A. I obviously computer vision and whatnot is changing that modality as well as what's going on in the home and whatnot.
So we're gonna have that webinar June 8th at one p. m. Easter time. We usually have it on the first Thursday. Happens to be a little too close to my anniversary. So we're going to do June 8th at 1 p. m. Eastern time future of care spaces. We would love to have you be a part of it. If you are interested in being there, go ahead and hit our website.
Top right hand corner. We have a card. You can click on that card and go ahead and fill out the form and get registered today. We would love to have you join us we look forward to seeing you there. Now back to our show.
Yeah, there's a lot of different things we can talk about here. I wanna go onto the Apple headset only because it'll be another six or seven weeks before you and I are on the air again. And this will have either happened or not happened because this has been talked about for a while.
But there's a belief that at this year's Apple Developer Conference, which will happen in I think early June, They're finally going to release their their VR platform. And Palmer Lucky is the person who came up with Oculus and sold it to meta. And he he said simply in a tweet, right? So it's just a tweet from Palmer.
Lucky. He said The Apple headset is so good. That's the quote. So, and some other people have been leaking things. And do you think now is the right time? Do you think Apple is going to release a headset and what do you expect to see from a headset from Apple?
Yeah, a couple of things.
So I'll give Amy Webb and the in the Futurist Institute some credit here. She has always said that, our future from devices and whatnot will be, you know what we have right here. That's what we'll be using. The this is gonna go away. There are several other pieces of technology that'll go away.
I think this, what we'll see from is the next step towards that. So it's not just virtual reality as you and I, think about, sitting at the beach and lounge chairs, having this type of discussion. And or, VR for gaming and so forth, even simulation. So it's gonna take it to the next level.
And I think what it's really gonna help us is enough people will adopt that we can get the requisite feedback and whatnot so that the next cycle of this is, improved and you can go back to, you remember, Google Glass and all the stuff that happened there. Oculus came out. This is more of a, kinda a combination of both technologies, if you will.
But we're gonna see this even advanced. No one's gonna wanna walk around with goggles on their head. That's not gonna happen. Folks like folks are okay with glasses, everybody uses, sunglasses and whatnot. So, more of the usability needs to be advanced going forward, but I think we're just gonna learn a lot from it.
Yeah, I, yeah. Even from Apple, I expect the first device to be a little clunky, and I agree with you. I don't think we're gonna see people walking down the street with their Apple headsets on just yet. But I think that is directionally where they're going. Where they're going is, yep. At some point we're gonna have our glass, I have my glasses on.
You have your glasses on. And even people who don't need glasses like we do, We'll have 'em on because they will, there will be augmented reality that goes on, and I agree with you. I think Apple sees a future in which that kind of device, which provides augmented reality also does all the things our current phone does.
Yeah. Wayfinding, all that good stuff, right?
Yep. Yeah. Maps, wayfinding searches on the internet. We could just do a voice search and it'll show up in front of us. Now I, I saw a presentation a little while back where they, they showed the the Oculus headset now is called the, what's meta thing called?
Anyway yeah the meta headset and they said, getting this technology into this, the simple pair of glasses is, still a decade off because you're talking about advanced this and advanced that. Oh, absolutely. Certain levelization and whatnot. But I think if we see anything from Apple, it'll be something where you and I sit at our desk.
We can still interact with our computer. We have augmented reality. And it will create a better experience for us to interact with our digital world while sitting here. And eventually we will be able to take that, that headset on the road with us. Yeah. And
Getting rid of this would be yeah, hands free would be something I'd look forward to.
And Charles, I keep this here as a reminder. I know we keep pointing to things. People on the podcast are like, what are you doing? So this doing, this is the original eight gig iPhone box. Now I wish I'd kept the actual iPhone cause it would be worth a bunch of money. But my gosh, when you look at this, it's, you think, how did we even function on it?
That thing was so small. To start with it, if you remember back, it didn't have I don't think it had an app store when the first iPhone came out. Right. So they have a tendency to, put something out. Get the feedback, come back, in and iterate to something great. And I think that's what they're gonna Yep.
They're gonna do here. And I agree
and Bill from about it, from what they've done from a health perspective and the whole 24 7. This even add to that. So, at some point in time in the future your health will be a hundred percent monitored. You'll be able to see where you are at any particular time.
You won't have to, you won't have to go to this or this. It'll be right out there in front of you.
Wow. How much money do you give to Apple a year? You got the EarPods in that you're talking to me on? You got the phone, you just picked up, you got the Apple Watch.
So, so Bill nothing but it goes back to.
My brothers and I have an apple, the first Apple one motherboard. We still have that. So, it's in a safety positive box, but started commerce with him in the seventies.
Man, that's awesome, Charles. It is always great to catch up with you and in the spirit of your shirt, Mahalo friend.
There you go.
And that is the news. If I were a CIO today, I think what I would do is I'd have every team member listening to a show just like this one, and trying to have conversations with them after the show about what they've learned.
and what we can apply to our health system. If you wanna support this week Health, one of the ways you can do that is you can recommend our channels to a peer or to one of your staff members. We have two channels this week, health Newsroom, and this week Health Conference. You can check them out anywhere you listen to podcasts, which is a lot of places apple, Google, , overcast, Spotify, you name it, you could find it there. You could also find us on. And of course you could go to our website this week, health.com, and we want to thank our new state partners again, a lot of 'em, and we appreciate their participation in this show.
Cedar Sinai Accelerator Clear Sense CrowdStrike, digital Scientists, optimum Pure Storage. Sure. Test Tao, site Lumion and VMware who have 📍 invested in our mission to develop the next generation of health leaders. Thanks for listening. That's all for now.