This transcription is provided by artificial intelligence. We believe in technology but understand that even the smartest robots can sometimes get speech recognition wrong.
Newsday: Epic Faces Lawsuits and the AI Proving Ground with This Week Health
[:I'm Bill Russell, creator of this week Health, where our mission is to transform healthcare one connection at a time. Welcome to Newsday, breaking Down the Health it headlines that matter most. Let's jump into the news.
Bill Russell: alright. It's Newsday and it's Sarah, it's me, and hopefully Drex joins, but, you know, he's late.
it's tough for him to start getting outta bed earlier again and all that stuff. I what? Oh, there he's,
Drex DeFord | This Week Health: Look, I show up and you're talking bad about
Bill Russell: no, I just, I just said. I just, I just said it might be hard for you to get outta bed early after the break. It was a good break.
Drex DeFord | This Week Health: It's very difficult.
Sarah Richardson: 5:00 AM is way overrated. I'm, I was doing seven on the, over the holidays and I was like, I was feeling like I'm gonna like being retired someday.
ast. We operate as a company [:Sarah Richardson: get up early.
Bill Russell: but you just gotta get up early.
And and my daughter is one of those people and it was kind of interesting to have her come east for the holidays and be here. becuase she, I, she gets up early becuase that's her habit. She gets up early. But
uh,
Drex DeFord | This Week Health: Yeah.
Bill Russell: so hey happy New Year to you guys.
Drex DeFord | This Week Health: day.
Sarah Richardson: Yay.
Bill Russell: And Drex, we can see over your backdrop now so that's new for the, that's new for the year.
Drex DeFord | This Week Health: It will depend on whether or not I'm standing up or sitting down. I'm doing a lot of PT this year, so I'm gonna like, I got some things I gotta fix and my PT guy says, you
Bill Russell: stand,
Drex DeFord | This Week Health: too much, you need to stand up more. So
Bill Russell: I wonder,
I remember that microphone. I've interviewed you using that microphone behind you.
Drex DeFord | This Week Health: Yes. This little guy right over here. Sure enough.
. I mean, it's just the same [:Sarah Richardson: Because it's my happy backdrop. It has all my Easter eggs in it, and it always makes me happy when I, stare at something all day, it should be something you wanna look at.
Bill Russell: Are you gonna have Jeremy redesign it this year or are you just gonna you'd like it it's sort of your brand now.
Sarah Richardson: It is my brand and only
Drex DeFord | This Week Health: yeah.
Sarah Richardson: me, because if you said, oh, you need to do something different, I mean, I could, I mean, if the 49 ERs actually win the Super Bowl, I may add that somewhere in this mix, but that hasn't happened since I was in
Bill Russell: I was gonna say in order for that to happen, don't they have to beat the Seahawks at some point?
Sarah Richardson: We'll have to beat the Eagles first.
Bill Russell: Ooh. Oh really? That's what they have in the first round.
Sarah Richardson: we'll already know if I'm changing my backdrop or not.
we touched on predictions for:So we have to talk about ai. I've already done a couple of interviews this year. I just, I talked to Chero yesterday and oh gosh, I did another interview. The, when we started last year, people were sort of apologetic. They would say, oh, I'm sorry, I brought up ai. I think that's gone by the wayside.
I think everyone recognizes like, this is uh, this is here to stay. And you know, board members are asking about it. They were probably asked about it over the Christmas break. Like, you know, remember when we used to go to Christmas break and we'd go to a family gathering and they'd say, Hey, can you fix my computer?
I think a lot of that has started to become like hey, do you have a good prompt? Like how do you get chat GPT to do this? And how do you get I had a couple of those conversations with uh, uh, people in o over the holidays and I think people have been storing it up. They're like, Hey are you using, I love this.
n your business? Like. And I [:So I think this is here to say it's pretty democratized. What are you guys seeing and feeling for AI for 2026 for healthcare?
Sarah Richardson: There are two things that I honestly have been leaning into is, and every conversation I've had with A CIO comes up and budget management. And it's not always like the doing more with perpetually less, which is true, but it comes back to how are we going to integrate automation and agentic usage of AI properties appropriately?
Whether that's creating your own LLMs from the morass of data that you have
em. That dovetails a hundred [:Drex DeFord | This Week Health: I commented on a post from a. Buddy of mine this morning and kind of his assertion was there's so many people who are just trying to figure out what it even means. Like if you ask 10 CISOs or 10 CIOs, what, you know, what's ai, what are you doing with ai? You get completely different answers and during city tour dinners, for example, we've heard people say things like, my, my CEO showed up and said, we just need to be doing more AI stuff not [00:06:00] defined don't understand what it is, like, just sprinkle some AI on it. We're behind. I just went to this conference and there's people doing all this AI stuff. So I think there's a couple of things. It's like the, you know, the blindfolded man, you know, people grabbing an elephant, one thinks it's a tree and one thinks it's a snake. Some of them just don't understand what it is or how to use it yet. Back to your point, bill earlier. the other one is the, think in the CEO example of just sprinkle some AI on it, there's a ready, fire, aim thing that's going on right now that me because. I think we've gone through this process when we were all CIOs Hobby shop departments building their own things and then it's not supported and we don't know about it, and it creates for risk from a cybersecurity or a privacy perspective. there's probably some of that going on right now too.
it's good. I really love it [:Bill Russell: My healthcare CIO playbook. The first of the 10 predictions was AI will reveal great leaders and expose others.
you know, I, I don't wanna put anything more on people than they already have it. They, but they have to learn ai. If you're in it, especially if you're in a CIO role. If you're a CTO role, CISO role you can't be saying.
I, you know, we don't have a good definition for it. Find one, make one up. you got to figure
Drex DeFord | This Week Health: it out.
Bill Russell: yeah. You've gotta figure this out. And it's really interesting. You know, one of the things I hear from people is like, well, we've gotta get our data right first. I'm like, yeah, well, okay. I've been hearing that for two decades.
ng, about definitions, about [:It's like, you know, we, AI doesn't understand length of stay. I am like, yeah, but your people. do And what we, what One of the things about AI is, and we're gonna hear this word a million times this year AI with context is amazing and we just marvel at how it responds. Like, oh my gosh, how did it know this much stuff?
AI without context, we think as an idiot, because if you don't give it context and it comes back and says, oh, you know, length of stay is this, you're like, oh, stupid thing Doesn't even understand what length of stay is.
Drex DeFord | This Week Health: It's Hallucinating.
Sarah Richardson: Here's the real question then. Do you believe that the leaders that hire CIOs have the wherewithal and knowledge to be able to have that conversation with the recruiting firms, to hire the right CIOs into their organizations to do exactly what you just shared?
Bill Russell: No But everybody in the industry knows my email address and they more, I'm more than happy to help any health system hire a good CIO and if they need a good ciso, I'm gonna have, you know,
Drex DeFord | This Week Health: same.
ll Russell: to drex and I, I [:Drex DeFord | This Week Health: The thing about ai, the other thing I, that I will point that I wanted to make was the idea that if you're not learning it, if you're not figuring it out now, like. Two things. One is in the process of figuring out AI and what it can be and what you can use it for that actually will help you find the problems with data and help you clean it up.
So that's a really good thing.
Bill Russell: we've experienced that internally, by the way, but go ahead.
Drex DeFord | This Week Health: Yep. This is, we're eating our own dog food right now on a lot of this stuff.
Sarah Richardson: champagne
Drex DeFord | This Week Health: Our own duplicate records, our own things that we yeah. Are doing. But the other thing about this, my brain works in analogies. I've been thinking about this as, and I'm sure I stole this from somebody, I didn't come up with it myself.
amazing things. The problem [:so get in there and start driving and figure it out.
Sarah Richardson: and to your point, Drex, even if you've, you know, seen the movie F1, becuase you're not as much, you know, if you're not as big of a fan, they're in the simulator for hours. And so whether you
Drex DeFord | This Week Health: Yeah.
Sarah Richardson: you've built or a lab you've borrowed from a partner, we have partners who have AI labs. You have to teach people how to use it in a safe environment so you don't go launch a bunch of stuff that goes rogue in your own environment.
Go learn how to do it.
Bill Russell: let's stop for one second. Did you see the movie F1.
Sarah Richardson: Yes.
Drex DeFord | This Week Health: yeah.
Bill Russell: Did you?
Drex DeFord | This Week Health: Yeah.
Bill Russell: I did too. I loved it. It was amazing.
Drex DeFord | This Week Health: Produced. Produced by Louis Hamilton. So
Sarah Richardson: so I mean
Bill Russell: I don't,
Sarah Richardson: becuase that.
Bill Russell: There was a couple thi a couple things. I watched all the the extras on it, and Louis Hamilton was a seven time champion
x DeFord | This Week Health: [:Bill Russell: and
Drex DeFord | This Week Health: Champion. Yep.
Bill Russell: Just just amazing. But he was, because they were looking for authenticity. He was watching the movie and he heard something. He goes.
That car's in the wrong gear that shouldn't be in that gear going around that turn. And they actually changed the sound to make none of us would've known, but they changed the sound to so that, that was pretty cool. But I here's what I took from that movie that I thought was pretty interesting.
One is they're in the wind tunnel. And by the way, those were all real places So those are all they got really incredible access, but they're looking at the wind tunnel and she's like, I'm trying to get you another 10th of a second.
Okay. 10th of a second is what, what, what she was looking for in the design and all this other stuff. I love that thinking. It's like, can we reduce access by one day
Sarah Richardson: Mm-hmm.
broken. It's wrong. It's the [:Bill Russell: I don't watch a lot of Formula one, and so I did not appreciate the fact that when that car's going around. I thought, well, it's the driver, it's the pit crew. And you know, away we go. They had like a whole call center full of people that were monitoring specific things. And in one of the races they call over and say, we need a sub three second pit.
Can we do it? And still remain in front And the person's like doing the calculations going, yeah, we can do it. Bring 'em in. I'm like, and it was a whole team of people and I thought, that's healthcare and it like.
temperature of the inside of [:That's how they get the, one more, 10th of a second. One more thousandth of a second.
Bill Russell: I look at some of these predictions out here. One of the things is staffing. People are always asking me about staffing with regard to ai. And, and, And Sarah you're, you keep bringing us back to this re-skilling and whatnot.
I I remember, when we were doing data governance at St. Joe's, we, at one point we went out and tried to find all the people that needed access to the data, right? This is one of the things you do early on. It is like, you know, are we gonna control this and bring 'em all into it? Or are we going to be sort of a, an educator and you know, a storehouse and whatnot?
people that we [:Bring in lots of information, analyze that information make sense of it, even generate reports from it. I mean, it does that like. In minutes. And you know, that's one of the areas where I'm like, okay if your title is analyst, you might want to figure out how to be the person who makes ai do what it should do, and not just rest on your laurels and say, well, they'll always need me.
k me also on the other side. [:It's like, you know, analyze this first, understand this first do this second, do this third, do this fourth. And those people are gonna be in really high demand. There was a post over the holidays. And Andrew Ning was did a thing where he used an old foundation model. With the right process flow.
And then he used a he did single shot on it, then he did the new model, so chat, GPT-4 chat, GPT five, single shot. And then he gave it a workflow, which is what I'm talking about here. These contact engineers do these workflows. He gave it the workflow and he used the old model.
g the initial foundation the [:Drex DeFord | This Week Health: about the context.
Bill Russell: about the context and it's about the person thinking through, like we would think through if I gave you the assignment to, you know, figure out what this week health and the 2 29 project should do next.
You would analyze the market, analyze the competition, analyze. I mean, you would have this flow of things that you would do. And instead we go to AI and we just say, Hey, what should we do next? And if it went through that flow of, analyze the competition, analyze the market, what do CIOs need, what do CISOs need, what do CTOs need?
It's going to give you a better answer. And I think those people are gonna be in high demand. What are you telling people as they say, Hey, how's this gonna impact our workforce? And what jobs are going to be needed and what jobs may not be needed?
lean in, and I think this is [:And then really digging into what can I build? So if I create like a rotational program as an example. That can develop people into specialized roles to fill some of these spaces. Knowing that access to national talent pools is gonna be huge becuase no one's gonna wanna move, in theory to a small Midwest town. I went to a big Midwest town and didn't still wanna live there. Having the right partners that have specialized expertise if you can't afford to hire them and they can afford to teach you, [00:18:00] and then the right narratives for the organization because people get really entrenched in, honestly, this is the way we've always done things and it's hard to move those needles.
And so when you realize it's going to be really hard to do this work and really worth it, you find that people that want to get into that space and then they get air cover. From honestly, the CEO has to set that time that goes back to change management 1 0 1. The person who's going to make it happen has to be the one sponsoring the fact that people are allowed to go do that, and it's gonna make people really uncomfortable for a while.
Drex DeFord | This Week Health: I, I talked about a lot of this in two minute drill yesterday, the ISC two, the ISC squared, report on cyber staffing kind of came out and a big part of it was just the, uh. need to know how to do this. People need to know how to do that. But the bottom line was there's not enough people.
th experience. And so one of [:And then the security that goes along with that. And a lot of those skills that those people have aren't necessarily, and they know how to build servers, they know how to run firewalls or all those things. They're people with skills like. They're really good collaborators and they understand how to problem solve and they, you know, they understand the workflows and so I think the people, back to Bill's point, the people who understand the workflows also have this great opportunity to get involved in AI now because the context comes from the workflow and the things that need to be done to get the work done. you can give that to the AI as context. It really helps you build an AI that can solve a lot of problems.
son: And we have really good [:Drex DeFord | This Week Health: Yeah.
Sarah Richardson: literally they will come and sit with you and watch you and say, yeah, this workflow is optimized. Now let's AI it.
Drex DeFord | This Week Health: Yeah.
Sarah Richardson: we have those conversations at our city tour dinners and summits today. But we already have those partners.
And if I was still in a hospital practicing today, I would be bringing those guys in to say, watch this group. Let's have a couple of pilots, not even pilots. Let's just beta test this, hold my beer, show the art of the possible.
Drex DeFord | This Week Health: Over the course of the next year, as we do more and more of our podcasts, the conversations that we're going to have, the angle of the conversations that we're gonna have is I think will be much more about. How did you solve that problem? What did you do? What was the thing that was the breakthrough moment for you, to solve that problem?
his is what everybody really [:Bill Russell: we're gonna close on this and I'm gonna give you a little bit of background. So there's three lawsuits against Epic Federal antitrust suit, brought by particle health claims. Epic unlawfully blocked competition. In the emerging payer platform market and leveraged its EHR Dominance to slow competitor's growth a case a judge has allowed it to proceed on monopolization claims.
ys that harm competition and [:Now I'm on record as saying the only problem with these monopolist claims is that they don't have a single client that is bringing a lawsuit against them. So typically when there's a monopoly it impacts the market and and what you're. The market complains. Here you have the government complaining and you have two competitors complaining.
It's like, Hey, they're a monopoly. They're a monopoly, but nobody on the other side is complaining of monopolistic pressures. I do think this is going to impact how Epic does business. I think they're going to be a little bit we, we can say that they've been heavy handed, would be the word I would use.
They have been heavy handed with their partners, especially, I've seen some of the contracts that they ask their partners to sign, and it's essentially, we can copy you, we can steal your stuff, we can steal your clients, and we owe you nothing.
t the largest clients in the [:They are maybe against partners, but they're not for the market. But I think one of the things we're gonna talk about a lot this year is the fact that others are. You know, 600% price increases. The price increases across the board, the SaaS models, the cloud providers. I think CIOs are gonna start complaining.
It's, it is gonna start with complaints, which we started to hear last year. And it's going to crescendo sometime this year. I don't know what it's gonna look like, but I think there's gonna be. A push for legislation of some kind that says, look they can't raise our price by 600% becuase they know we're gonna pay it.
I mean, it's like,
Drex DeFord | This Week Health: Price controls.
d so I know there's a lot of [:But I,
Drex DeFord | This Week Health: had, you just did the show with Chuck
Bill Russell: yeah.
Drex DeFord | This Week Health: Talking about the. one of his cloud providers kind of pulling the rug out from under him or something. Right. And then having to hurry up and figure out what his backup plan was. And then that always kind of puts you in a situation where it's like, well, it's an emergency.
I guess I'll pay whatever it is that you tell me that I have to pay. That's the other situation, and I think a lot of us find ourselves in, especially in the. Heat of the moment. You know, I always, of course I'm gonna think about this as a security thing, but when a security event happens and you need help, these are the dudes with fire trucks in the water, we're gonna pay 'em whatever it is that we need to pay 'em to put the fire out.
I think there's that part of it too.
in front of you that's, you [:Sarah Richardson: You start to find alternatives for small pieces of it, and I say this with all due respect, that I'm just gonna throw 2030 out there for no particular reason. You will have a way to replace a lot of these things through your own volition because of what AI will bring to the table, and there will be all the functions of an EMR built around own centralized space.
ee pieces of it get replaced [:Bill Russell: Hey, that's, we should start a practice.
Sarah Richardson: back in a hospital. But that's how it works.
And you get
Who knows how to do it. One person who gets people to wanna do it, one person makes sure it's safe to do it. There you go. And
Drex DeFord | This Week Health: It's an interesting idea to, it's a really interesting kind of concept that the, you know, we used, we a lot of, this is just a historical thing. We bought a whole bunch of different applications from lots of different companies and we tried to figure out how to strap 'em together through an interface engine, and it sucked.
orks together. That's really [:And we're back to, there's a whole bunch of stuff because of the tech that we have now with AI and APIs and lots of other stuff. A lot of these little things could be built, it could be really customized for transplant or for the emergency department. It could work amazing and it, we love it because of the workflow and it's gonna be able to integrate well. know. Yeah, I think it's an interesting idea. There could be a lot of, could be a whole new world here in the next 18 months.
Sarah Richardson: you never need to retire. You can just keep ideating
Drex DeFord | This Week Health: Keep going.
Sarah Richardson: me and.
Bill Russell: You guys are scaring me now. Yeah, I think the EHR providers would say that they're a combination of a handful of things they're gonna say. The data model becuase Epic did have their data model stolen by somebody in India, I remember back in the day.
the workflow, although it's [:It's gonna be the ecosystem that they've created. And the integrations that they've created are going to be all those things. The hard things to replicate is just that. It is, it is the scale at which they've done things. It's sharing things across an entire ecosystem and whatnot. But the thing is, I really wish somebody would go in this direction.
I don't think they're going to. Is you know, essentially open it up so that the user interface could be spoken into existence. That we could vibe code the user interface on top of these things.
Drex DeFord | This Week Health: Mm-hmm.
Bill Russell: then we're talking about a level of customization for the, what you were talking about, Sarah, for the clinicians and the physicians that is at a different level where it's like this is specifically designed.
For Sarah Richardson, O-B-G-Y-N, who practices in Kansas City and you know, and it, I'm sorry, I put you in Kansas City.
Sarah Richardson: Yeah, I moved.
Bill Russell: [:I, I could see a future like that, but it would require the EHR providers and others to think very differently about that.
Drex DeFord | This Week Health: are changing and the tech is changing, right? So that thing that you just described, I was a Cerner client, they had a thing called mage. And Mage could be used to build exactly what you wanted, and it ran on the Cerner database, the EHR database, and ultimately it just. You know, generally rejected.
It was hard to, you know, harder to build in pages and you had to keep 'em updated and then there were new releases and blah, blah, blah. I just think that we're, the tech is coming along at such a speed that I don't know that problem is going to be a problem in the future. So,
Bill Russell: Yeah.
| This Week Health: here we [:Sarah Richardson: Voice direction. No keyboards. The most
Drex DeFord | This Week Health: No keyboards.
Sarah Richardson: no keyboard.
Bill Russell: Yep. I don't know that I don't know. I'd love to tell you if Epic's going to win or lose any of these lawsuits, I don't think they'll lose the particle health one, but the damage will be done, right. So, they will do discovery. There will be stuff made public that Epic may not wanna be made public.
They will interview people become part of the record. And that's you know, it's just all this, the reason I think it'll impact how they do business is not because they're gonna lose they could theoretically win all of these lawsuits. They just don't want to keep having lawsuits.
Is what's gonna change how they how they approach business moving forward. So, wow. It is great to be back with you guys. I mean, not that I don't enjoy our breaks and time off, but it is great. I look forward to seeing you guys next week. In person in beautiful Silicon Valley. Should be fun and thanks for listening.
That's all for now.
th our Daily Insights email. [: