theCUBE + NYSE Wired: MedTech Unplugged, the Future of AI in Healthcare & Life Sciences | Blake Evans, Rush University System for Health & David Unger, Adonis

Clips
More from theCUBE + NYSE Wired: MedTech Unplugged, the Future of AI in Healthcare & Life Sciences

Gemma Allen

Host, theCUBE + NYSE Wired

theCUBE

Blake Evans

System Vice President,

Rush University System For Health

David Unger

Managing Director, Health Systems Sales

Adonis

play_circle_outline Rush University System: revenue cycle leadership, 700-person team, patient financial advocacy

play_circle_outline Adonis AI-Native Decisioning and Work-by-Exception Automation for Revenue Cycle

play_circle_outline Vendor lock-in and frontier models: Epic/ChatGPT plugins versus specialized vendors

play_circle_outline Automation vs Intelligence in Healthcare Revenue Cycle AI: Billing, Claims, Denials, High-Value Appeal Decisioning

play_circle_outline AI-Driven Revenue Cycle Automation and Collections with Decisioning to Prioritize High-Dollar Appeals

play_circle_outline Industry outlook: providers and payers adopting AI; potential AI-vs-AI dynamics.

play_circle_outline Voice AI agents and the prospect of AI-on-AI interactions between payer and provider

play_circle_outline Data ingestion challenges: EHRs, payer contracts, policies, structured versus unstructured data.

play_circle_outline Data readiness: ingesting EHRs, payer policies, contracts, and structuring data

play_circle_outline Vendor lock-in concerns and frontier AI models like ChatGPT and Claude in healthcare.

Info

Blake Evans, Rush University System for Health & David Unger, Adonis

Gemma Allen

Host, theCUBE + NYSE Wired theCUBE

HOST

Blake Evans

System Vice President, Rush University System For Health

David Unger

Managing Director, Health Systems Sales Adonis

search

Blake Evans

>> Palo Alto Studio Connection, Silicon Valley and Wall Street. I'm John Furrier, hosting theCUBE here with Dave Vellante, my co-host.

Gemma Allen

>> Welcome back to theCUBE studio here at the New York Stock Exchange. I'm Gemma Allen, co-host of NYSE Wired, MedTech Unplugged. And today we are having a conversation about AI in healthcare, how it usually conjures robosurgeons, and miracle drug discovery. But there's another enormous problem technology is coming for. The bill. American hospitals deliver billions of dollars of care every year. They navigate an extraordinary maze of insurers, claims, denials, and collections to actually get paid. So what happens when AI starts doing the chasing? Joining me now is Blake Evans, System Vice President of Revenue Cycle at Rush University System for Health, and David Unger, Managing Director of Health Systems at Adonis. Welcome, folks.

David Unger

>> Thank you. Thank you for having us.

David Unger

>> Thanks for having us.

Gemma Allen

>> So first, to start, I think I laid it out there. The bill is not something that we think about perhaps top of mind when we think about the world of AI in healthcare, but it's a very important problem to solve, right? Especially in the world of patient experience, of economic efficiency, all of which AI has promised. Both of you are seasoned in this industry. You've been in the healthcare space a long time, coming at it from different parts. So to begin, maybe just start with you, Blake. Introduce yourself. Tell me a little bit about what it's like to solve for a problem like this in Chicago with the healthcare remit that you lead.

Blake Evans

>> Yeah, so, very happy to be here. I'm Blake Evans. I'm the Assistant Vice President of Revenue Cycle at Rush University System for Health. I lead a team of over 700 individuals that are really working to help advocate for our patients, get our billing done. I always call revenue cycle the lifeblood of the organization because we are everything from the point of that initial patient contact as it relates to registration all the way through that billing and collections. So if you think about the non-clinical experience for that patient, everything financial is going to be resonating in that revenue cycle space.

Gemma Allen

>> And David Unger; Adonis, interesting company at a very interesting time. AI native, AI led. Talk me through the company, the mission, and where you guys, I guess, are seeing opportunity in the market.

David Unger

>> Yeah, absolutely. Thanks again, too, for having us. So Adonis, we believe that today's revenue cycle is broken. We think that there is far too much manual labor. Not enough intelligence happening to figure out really what are the things that our teams should be focused on. So Adonis was formed to really make revenue cycle flow much more autonomously. And we like to think that working by exception is the future. So we're focused on how do we take all of the data that floats around within a health system, ingesting that as an intelligence layer and giving organizations the insights on where they should focus And then also from an agentic automation standpoint, taking care of the work that can be done by agents. So that's our focus. And again, we're partnered with health systems and provider groups across the US.

Gemma Allen

>> So let's talk about the trillion-dollar elephant in the room, right? Because it is a word we say or a topic we talk about maybe 50 times a day on this show. It is everything right now. Talk about the AI moment, though, from the perspective of healthcare and automation and efficiency, because, I mean, I only know as a patient what it's like to ring an insurer, right? It is painful. you go in the loop, you talk to get to the operator, you talk to somebody, you repeat your credentials 300 times. I can imagine from the perspective of a hospital network, it's equally as taxing.

Gemma Allen

>> Yeah.

Gemma Allen

>> Talk me through what you're seeing and how quickly, I guess, and efficiently AI can actually create a solve for that mundane challenge that many of us face personally and professionally?

Blake Evans

>> Yeah, so it's a great question. I think in revenue cycle, if you think about all the transactions that we deal with every single day, every single patient that comes through our organization, we not only have to touch them at one point of their experience, but all the way through, right? So in the revenue cycle, you're likely to touch that same patient 4 or 5 times at various stages. So it'd be at registration or coding, or billing or collections through that workflow or that cycle of revenue cycle. And so if you think about the core of that, there are thousands and thousands and thousands of transactions that happen every single day. And so most times in revenue cycle, longstanding history is that we've been taught to complete the transaction. We've been taught to complete the next step. But where AI is really giving us an opportunity is to say, what do we really have to do versus things that the technology can help do for us, right? To make things more efficient, to make it more— the timeline speed up. And so why you see so much attention in the AI space as it relates to healthcare and also in revenue cycle is because of that transaction-based mentality and how can we turn those transactions into more of an automated workflow.

Gemma Allen

>> And for you, at Adonis, on the team there, how do you think about proving that efficiency? what is the measure by which you can say, We can speed this up for you. We can make this more cost effective for you. We can save costs here, here, and here. How do you present that opportunity?

David Unger

>> Yeah, I think, with AI, I think it sometimes gets used as a universal term when in reality a lot of times what we're talking about is AI as an intelligence, a decisioning layer versus automation, which is kind of a separate function. And so I think to Blake's point, We do both of them and we're really focused on, again, where there's high volume, high frequency. Let's apply automation to that work because it's not really something that a human needs to be spending time on. And so we kind of say, okay, high volume, push that more into the automation bucket. But then as you're thinking about the more highly complex, perhaps lower volume, that's much more a question of decisioning and intelligence and what's the right thing that we need to do as an organization for a health system to be paid. And I think appeals are a really great example where a high-dollar appeal can sometimes be hundreds of thousands, if not millions of dollars for a health system. And so, the decisioning to know what are the appeals to actually go after and spend time on, what to send to the insurance companies that's going to increase your likelihood of that appeal being overturned, that's where we see AI playing a really big piece. And so, I think when you think about how do you prove that, you first need to understand what's your goal for that set of AR. But then certainly from efficiency standpoint, we're able to do the work of hundreds of humans, if you will, with automation. So there's a direct connection to decreasing cost to collect for organizations.

Gemma Allen

>> So I saw this concept of exceptions, right, in revenue exceptions in your product suite. Talk me through that. Maybe help me understand. So you're like working in a billing department of a large hospital. You have hundreds of claims that you're working through. This will essentially— this AI will help you understand where the rubber meets the road from the perspective of what actually needs human intervention versus what could potentially be managed by an agent or in some sort of automated fashion. In the latter option, what happens then? Does an agent talk to an agent at an insurer? how is this product and technology evolving within the workflow? So I'll ask you that, Blake.

Blake Evans

>> Yes, it's a great question. I think there's various use cases for automation or AI, to David's point, I think, what we're seeing more fundamentally is that this is a transformation of our workforce, a transformational time of really the industry, right? And so I think this whole concept of, we're going to get to a point where AI is fighting AI, the AI on the provider side is fighting AI on the payer side. But what I really think is that the AI should be used not to say what's all the things that we can automate, right? But we should start to think, okay, what are the tasks that we should not be doing anyways, right? And how do we use the data that's been infused in the AI to become more intelligent and make things more efficient? So at the end of the day, I do think that the payers are going to invest in AI, the providers are going to continue to do that. But it's more about how do we make sure that our patients are taken care of most efficiently? How do we get their bills processed more efficiently and how do we communicate to them so that they know any outstanding responsibility and kind of have a good financial outcome?

David Unger

>> Yeah.

Gemma Allen

>> When we talk to enterprise customers on this show, even Fortune 500 companies, we hear a lot about the promise of AI, right? But at the end of the day, a lot of them have had a level of vendor— I don't want to call it a hostage situation, but close enough to— with certain tech stacks for a number of years. Right. That it's very difficult to unlock. It's difficult to remove line of business applications that have been in place for a very long time. I'm going to put this to you, David. How easy is it for a healthcare system to adopt a technology like Adonis? how quickly can you plug and play or deploy this? what's involved?

David Unger

>> Yeah, great question. I think the adoption piece, again, it depends on the scope of the AI. And so if you're looking for intelligence decision framework, I would say the disruption is less because you're not necessarily changing core workflows. You're not having to do complicated tech stack integrations. You're simply bringing forward more information about what's the right thing to do for your business. When you start looking at building more of the direct automation. So you mentioned voice AI calling earlier. So that certainly is a thing where health systems will have voice AI agents talking to voice AI agents on the payer side. That integration does take a little bit more time because there's still a lot of technical complexity that goes into it. But ultimately, I think adoption is always going to come down to the leaders. A great leader like Blake is going to make sure that his team understands what is the technology that he's bringing in, why is he bringing it in, what is the value that it's going to derive. And I think as long as the team members are bought into the vision of the leadership, you can get to good adoption. But just from a timeframe standpoint, I would say you can reasonably adopt AI technologies in rev cycle. Within 3 to 6 months from the launch

Gemma Allen

>> date.Yeah, that's a reasonable timeframe too, right? We hear people say tomorrow and I'm always a little bit

David Unger

>> skeptical.

Gemma Allen

>> Yeah.So I actually love to hear answers like that. So Blake, you mentioned you have 700 people in your organization. I'm sure there's a lot of very committed and very loyal folks there. Talk to me about the human element to this. interestingly, one thing we hear all the time in the news is that the one sector that's growing a lot within the jobs report in America right now is healthcare, right? Perhaps that's clinicians. I'm not sure. How do you see this playing out from the perspective of technologies like this implemented within your stack and what would it mean for your workforce? How are people responding internally to these conversations?

Blake Evans

>> It's a great question. I think one of the risks of all this automation and AI is the mindset would go is that you immediately can get rid of people or you no longer need these people to do certain tasks or certain jobs. And so we have taken a very different approach at Rush. We have really, one, set the vision that we are going to embrace automation, we're going to embrace AI, right? It's not going away, right? And we use it as a way to get rid of menial tasks for our workforce, right? And not only that, but when we are asking for where can we automate or what can we do differently, we ask our frontline staff that, right? A lot of times they are the people who know this takes me hours to do, but it really doesn't add a ton of value. It's just a step I have to take. And so they have become our greatest source of, you know, where do we go, where do we focus, right? And they have now become really great advocates of where do we go next, right? Which has been really, really important. I think the other opportunity here is that as a leader, as David mentioned, you have to, one, help your team embrace the idea of using the automation. But also, if you tell somebody, I'm going to take away 30% of your work, and you still have 70%, if you don't tell them what to do with that other 30% of their time, that can create anxiety for them 'cause they're saying, "My job is now going away." But we want to instill confidence in our team, "This is how we can use it, and here's what it'll allow you to do differently and what more we can do with the technology." And so we've taken that principle in terms of, one, making it a part of our vision at Rush and what we do and how we deploy our team. Two, we get our staff involved in that process. And then three, we talk about the outcomes. We talk about when it works. We talk about when it doesn't work. And when it does work, we also say, "Hey," You see that that didn't result in anybody losing their job. So we actually talk about that with our team so it can help ease some of that tension and have them really think through how do we evolve our workforce so they can embrace it more and more and more.

Gemma Allen

>> Staying on the topic of readiness for a second, I want to ask you both about data readiness within this deployment and this opportunity for automation? Again, we know that that has been a challenge for many industries in many spaces for many decades. How is it changing within this world of AI and automation and efficiency? And, if you're working in a billing department and you refer to something as underpaid and somebody else refers to something as, it's a different even like, taxonomy, right? That can create a whole lot of challenges in the world of structured data that's needed to feed, an automated feature like this.

Blake Evans

>> Yeah.

Gemma Allen

>> Talk me through how— what you're seeing more so and how— I'm asking you both this. How you're kind of responding to that from a solutions perspective.

Blake Evans

>> Yeah, one of the limitations— and David, I want to hear what you got to say too— but one of the limitations that we've had from a provider side is that we only know our own kind of data, our own environment. You mentioned underpayments, right? There's so much that goes into that. One is the contract that you negotiated with the payer, right? A lot of times we don't have access to that contract, but where AI and automation is really starting to help us is you can ingest different data sources so you can pull in payer policy, you can pull in your contract with the payer, you can pull in coding guidelines, right, to make your staff a lot more intelligent. So when we are trying to fight a denial or we say, hey, this is underpaid, we have the actual background in the contract and how much and where we found it in the clause. And that's the level of insight that the technology is starting to give us where we would not have had that historically.

Gemma Allen

>> And for you, David, when you're rolling out this product, Adonis products, to healthcare institutions across the US. How is that problem being managed? How front and center is this idea of structured data versus unstructured into an ingestion engine? But what do you think happened there?

David Unger

>> Yeah, it's a good question. I think, Blake, you hit on a big piece, which is data is going to be king. I don't think that's a new thing necessarily. But when you think about intelligent decisioning, whoever has the most data, the most information is going to be able to create the best insights. So data is very important. To your question about what's the process to structure it, when you think back to the adoption, 3 to 6 months, that's a big part of what's happening there is Rush is different than other health systems within its own region, certainly different than other health systems in, call it, the East Coast or the West Coast. So there certainly is still, I would say, a bit of work that needs to go in to make sure that the data fits the model. Having said that, again, there are kind of a finite number of data sources within healthcare, the EHR being the primary one. And so that's where we currently focus most of our efforts is how do we best understand the EHRs like Epic, like athenahealth, like Cerner and Oracle Health. And then moving into, as Blake mentioned, insurance policies, contract data, ERP data. So I think it's taking a structured approach to get the most important data right first and then you expand from there.

Gemma Allen

>> So moving on to the last question I have for you guys, maybe the spicier question on the pod here today. The world of AI frontier models, it's moving fast, right? We heard Claude for Healthcare is making a big push into the industry. We know ChatGPT has just released an agent plugin for Epic. It's interesting when we're moving from data to that because some folks have concerns around data retention and what that really means if you have one core orchestrator across your entire business, right? There is a dependency there for sure.

Gemma Allen

>> Mm-hmm. How—

Gemma Allen

>> and I'm going to maybe start with you, David. How do you think about that from a competitive perspective? When you see these news announcements from these big frontier models, you're like, oh God, or are you like, well, actually, yeah, give me a response.

David Unger

>> Yeah, it's— we certainly pay attention to them. I would say, though, the world of healthcare is still very, very complex. And so I think to believe that you can just install a plugin or you can take some AI features or functionality off the shelf is probably not going to be the case. Again, the way Rush and Blake operate as a health system is specific to their payer mix, to their patient population, to the demographics within that area. And so there it's not as easy as just putting an integrator in. And so as much as we believe AI and automation is going to just make things easy, there's still a lot of work that has to happen around there. So that's where we believe Adonis is fully focused within healthcare revenue cycle. It's all we do backend specifically because we believe, again, in order for AI to be useful, it has to be specific to your organization. And so I, again, we pay attention to the big names, the Epics and the ChatGPTs of the world, but we believe what we're building at Adonis is really going to move the needle and provide a lot of value for customers like Blake.

Gemma Allen

>> And Blake, I mentioned earlier this whole concept of vendor lock-in, right, which has definitely been a thing in enterprise and I'm sure in healthcare and public sector too for quite a while. We've had an interesting couple of weeks in the news. We've had all sorts of reports of, you know, there might be one orchestrator across the world, actually not just across any one enterprise and no names mentioned. But certainly it's an interesting dilemma, right? Because in some respects it's another version of, the '90s and what kind of— what about that? Talk about healthcare though. How realistic do you think that is? How realistic do you think it is that you would have one company kind of managing everything for you?

Blake Evans

>> Yeah, I doubt that that will come to fruition. Right. I think that the complexities of healthcare, not everybody can do everything well. Right. So that's the other challenge. When you look across the AI automation platform, there are thousands of companies now that are investing in AI, right? So as a healthcare provider, I have all of these options, right? I have my own EHR system, right, that can do some automation. I have these vendors that are kind of talking to me every single day about what they can do. And so I think what we have learned is how do we evaluate based off of the needs that we have? What are the problems that we're trying to solve? That's one. Two, we do short pilots, right, to really evaluate, make sure we get the full benefit of the product before we move forward in a broader capacity. And then 3, again, I'll just kind of circle back. It's important to get people involved. My frontline staff is going to have a totally different level of visibility into how they can use the automation versus my directors or myself as the top leader.

Gemma Allen

>> Okay.Yeah. Okay. So, gents, I know you're in New York from the Midwest for a couple of days. I want to talk about the next few days. What's ahead for you both professionally? What does the next— I mean, it's hard to think beyond 6 months in this world of tech right now, but Let's start with you, Blake. Well, what are the priorities between now and I guess 2027?

Blake Evans

>> Yeah. So one of our big focuses is really focusing back on the people, right? We want to really develop our workforce. We're looking at AI literacy for our team. One, how do we prepare them for the work? How do we prepare them for the future? How do we make them more analytical and problem solvers and looking at exceptions versus just looking at the mundane tasks every single day? So that's one big area of focus. And then two, We want more orchestrated AI, right? More intelligent AI versus just agentic, meaning it does a specific task. We're really trying to learn from the data and insights and become more intelligent.

Gemma Allen

>> And for you and the team at Adonis, what's ahead? You guys have a few new features coming out. You guys raising more money? Talk me through what the roadmap looks like.

David Unger

>> Yeah, so at Adonis, we recently closed our Series C just a couple of months ago. So not immediately looking to raise more money. But from a future standpoint, I agree with Blake where we're investing deeper and deeper into intelligence and decisioning. So again, how can we give our clients the best information, knowing what to work, when to work it, how to work it? And I would say the other piece we're focused on is ROI. So how do we make sure that our clients, can see and measure and understand the value of AI and automation? And that's a hard piece. Quite frankly, we talked a little bit about this earlier where ROI, when you adopt a new technology, takes a little while to figure out how do you quantify that. But I do think there's a cost of doing nothing. And so we certainly think that some of the early adopters of AI and technology, although they may not know exact ROI day one, if you fast forward 12, 18 months, I think those are going to be the organizations that are seeing major wins and successes because they took a chance and they started to do something versus wait. So those are our two core focuses. How do we do better intelligent decisioning and how do we better articulate ROI for our clients?

Gemma Allen

>> Well, that's certainly an interesting thought to leave with— the cost of doing nothing in 2026. What exactly is it? Gents, thank you so much for joining us at NYSE Wired in theCUBE.

David Unger

>> Thank you for having us.

Gemma Allen

>> I'm Gemma Allen here at theCUBE studio at the New York Stock Exchange. This is MedTech Unplugged, one of our programs at NYSE Wired. Thanks for watching.

Blake Evans, Rush University System for Health & David Unger, Adonis

search

Blake Evans

>> Palo Alto Studio Connection, Silicon Valley and Wall Street. I'm John Furrier, hosting theCUBE here with Dave Vellante, my co-host.

Gemma Allen

>> Welcome back to theCUBE studio here at the New York Stock Exchange. I'm Gemma Allen, co-host of NYSE Wired, MedTech Unplugged. And today we are having a conversation about AI in healthcare, how it usually conjures robosurgeons, and miracle drug discovery. But there's another enormous problem technology is coming for. The bill. American hospitals deliver billions of dollars of care every year. They navigate an extraordinary maze of insurers, claims, denials, and collections to actually get paid. So what happens when AI starts doing the chasing? Joining me now is Blake Evans, System Vice President of Revenue Cycle at Rush University System for Health, and David Unger, Managing Director of Health Systems at Adonis. Welcome, folks.

David Unger

>> Thank you. Thank you for having us.

David Unger

>> Thanks for having us.

Gemma Allen

>> So first, to start, I think I laid it out there. The bill is not something that we think about perhaps top of mind when we think about the world of AI in healthcare, but it's a very important problem to solve, right? Especially in the world of patient experience, of economic efficiency, all of which AI has promised. Both of you are seasoned in this industry. You've been in the healthcare space a long time, coming at it from different parts. So to begin, maybe just start with you, Blake. Introduce yourself. Tell me a little bit about what it's like to solve for a problem like this in Chicago with the healthcare remit that you lead.

Blake Evans

>> Yeah, so, very happy to be here. I'm Blake Evans. I'm the Assistant Vice President of Revenue Cycle at Rush University System for Health. I lead a team of over 700 individuals that are really working to help advocate for our patients, get our billing done. I always call revenue cycle the lifeblood of the organization because we are everything from the point of that initial patient contact as it relates to registration all the way through that billing and collections. So if you think about the non-clinical experience for that patient, everything financial is going to be resonating in that revenue cycle space.

Gemma Allen

>> And David Unger; Adonis, interesting company at a very interesting time. AI native, AI led. Talk me through the company, the mission, and where you guys, I guess, are seeing opportunity in the market.

David Unger

>> Yeah, absolutely. Thanks again, too, for having us. So Adonis, we believe that today's revenue cycle is broken. We think that there is far too much manual labor. Not enough intelligence happening to figure out really what are the things that our teams should be focused on. So Adonis was formed to really make revenue cycle flow much more autonomously. And we like to think that working by exception is the future. So we're focused on how do we take all of the data that floats around within a health system, ingesting that as an intelligence layer and giving organizations the insights on where they should focus And then also from an agentic automation standpoint, taking care of the work that can be done by agents. So that's our focus. And again, we're partnered with health systems and provider groups across the US.

Gemma Allen

>> So let's talk about the trillion-dollar elephant in the room, right? Because it is a word we say or a topic we talk about maybe 50 times a day on this show. It is everything right now. Talk about the AI moment, though, from the perspective of healthcare and automation and efficiency, because, I mean, I only know as a patient what it's like to ring an insurer, right? It is painful. you go in the loop, you talk to get to the operator, you talk to somebody, you repeat your credentials 300 times. I can imagine from the perspective of a hospital network, it's equally as taxing.

Gemma Allen

>> Yeah.

Gemma Allen

>> Talk me through what you're seeing and how quickly, I guess, and efficiently AI can actually create a solve for that mundane challenge that many of us face personally and professionally?

Blake Evans

>> Yeah, so it's a great question. I think in revenue cycle, if you think about all the transactions that we deal with every single day, every single patient that comes through our organization, we not only have to touch them at one point of their experience, but all the way through, right? So in the revenue cycle, you're likely to touch that same patient 4 or 5 times at various stages. So it'd be at registration or coding, or billing or collections through that workflow or that cycle of revenue cycle. And so if you think about the core of that, there are thousands and thousands and thousands of transactions that happen every single day. And so most times in revenue cycle, longstanding history is that we've been taught to complete the transaction. We've been taught to complete the next step. But where AI is really giving us an opportunity is to say, what do we really have to do versus things that the technology can help do for us, right? To make things more efficient, to make it more— the timeline speed up. And so why you see so much attention in the AI space as it relates to healthcare and also in revenue cycle is because of that transaction-based mentality and how can we turn those transactions into more of an automated workflow.

Gemma Allen

>> And for you, at Adonis, on the team there, how do you think about proving that efficiency? what is the measure by which you can say, We can speed this up for you. We can make this more cost effective for you. We can save costs here, here, and here. How do you present that opportunity?

David Unger

>> Yeah, I think, with AI, I think it sometimes gets used as a universal term when in reality a lot of times what we're talking about is AI as an intelligence, a decisioning layer versus automation, which is kind of a separate function. And so I think to Blake's point, We do both of them and we're really focused on, again, where there's high volume, high frequency. Let's apply automation to that work because it's not really something that a human needs to be spending time on. And so we kind of say, okay, high volume, push that more into the automation bucket. But then as you're thinking about the more highly complex, perhaps lower volume, that's much more a question of decisioning and intelligence and what's the right thing that we need to do as an organization for a health system to be paid. And I think appeals are a really great example where a high-dollar appeal can sometimes be hundreds of thousands, if not millions of dollars for a health system. And so, the decisioning to know what are the appeals to actually go after and spend time on, what to send to the insurance companies that's going to increase your likelihood of that appeal being overturned, that's where we see AI playing a really big piece. And so, I think when you think about how do you prove that, you first need to understand what's your goal for that set of AR. But then certainly from efficiency standpoint, we're able to do the work of hundreds of humans, if you will, with automation. So there's a direct connection to decreasing cost to collect for organizations.

Gemma Allen

>> So I saw this concept of exceptions, right, in revenue exceptions in your product suite. Talk me through that. Maybe help me understand. So you're like working in a billing department of a large hospital. You have hundreds of claims that you're working through. This will essentially— this AI will help you understand where the rubber meets the road from the perspective of what actually needs human intervention versus what could potentially be managed by an agent or in some sort of automated fashion. In the latter option, what happens then? Does an agent talk to an agent at an insurer? how is this product and technology evolving within the workflow? So I'll ask you that, Blake.

Blake Evans

>> Yes, it's a great question. I think there's various use cases for automation or AI, to David's point, I think, what we're seeing more fundamentally is that this is a transformation of our workforce, a transformational time of really the industry, right? And so I think this whole concept of, we're going to get to a point where AI is fighting AI, the AI on the provider side is fighting AI on the payer side. But what I really think is that the AI should be used not to say what's all the things that we can automate, right? But we should start to think, okay, what are the tasks that we should not be doing anyways, right? And how do we use the data that's been infused in the AI to become more intelligent and make things more efficient? So at the end of the day, I do think that the payers are going to invest in AI, the providers are going to continue to do that. But it's more about how do we make sure that our patients are taken care of most efficiently? How do we get their bills processed more efficiently and how do we communicate to them so that they know any outstanding responsibility and kind of have a good financial outcome?

David Unger

>> Yeah.

Gemma Allen

>> When we talk to enterprise customers on this show, even Fortune 500 companies, we hear a lot about the promise of AI, right? But at the end of the day, a lot of them have had a level of vendor— I don't want to call it a hostage situation, but close enough to— with certain tech stacks for a number of years. Right. That it's very difficult to unlock. It's difficult to remove line of business applications that have been in place for a very long time. I'm going to put this to you, David. How easy is it for a healthcare system to adopt a technology like Adonis? how quickly can you plug and play or deploy this? what's involved?

David Unger

>> Yeah, great question. I think the adoption piece, again, it depends on the scope of the AI. And so if you're looking for intelligence decision framework, I would say the disruption is less because you're not necessarily changing core workflows. You're not having to do complicated tech stack integrations. You're simply bringing forward more information about what's the right thing to do for your business. When you start looking at building more of the direct automation. So you mentioned voice AI calling earlier. So that certainly is a thing where health systems will have voice AI agents talking to voice AI agents on the payer side. That integration does take a little bit more time because there's still a lot of technical complexity that goes into it. But ultimately, I think adoption is always going to come down to the leaders. A great leader like Blake is going to make sure that his team understands what is the technology that he's bringing in, why is he bringing it in, what is the value that it's going to derive. And I think as long as the team members are bought into the vision of the leadership, you can get to good adoption. But just from a timeframe standpoint, I would say you can reasonably adopt AI technologies in rev cycle. Within 3 to 6 months from the launch

Gemma Allen

>> date.Yeah, that's a reasonable timeframe too, right? We hear people say tomorrow and I'm always a little bit

David Unger

>> skeptical.

Gemma Allen

>> Yeah.So I actually love to hear answers like that. So Blake, you mentioned you have 700 people in your organization. I'm sure there's a lot of very committed and very loyal folks there. Talk to me about the human element to this. interestingly, one thing we hear all the time in the news is that the one sector that's growing a lot within the jobs report in America right now is healthcare, right? Perhaps that's clinicians. I'm not sure. How do you see this playing out from the perspective of technologies like this implemented within your stack and what would it mean for your workforce? How are people responding internally to these conversations?

Blake Evans

>> It's a great question. I think one of the risks of all this automation and AI is the mindset would go is that you immediately can get rid of people or you no longer need these people to do certain tasks or certain jobs. And so we have taken a very different approach at Rush. We have really, one, set the vision that we are going to embrace automation, we're going to embrace AI, right? It's not going away, right? And we use it as a way to get rid of menial tasks for our workforce, right? And not only that, but when we are asking for where can we automate or what can we do differently, we ask our frontline staff that, right? A lot of times they are the people who know this takes me hours to do, but it really doesn't add a ton of value. It's just a step I have to take. And so they have become our greatest source of, you know, where do we go, where do we focus, right? And they have now become really great advocates of where do we go next, right? Which has been really, really important. I think the other opportunity here is that as a leader, as David mentioned, you have to, one, help your team embrace the idea of using the automation. But also, if you tell somebody, I'm going to take away 30% of your work, and you still have 70%, if you don't tell them what to do with that other 30% of their time, that can create anxiety for them 'cause they're saying, "My job is now going away." But we want to instill confidence in our team, "This is how we can use it, and here's what it'll allow you to do differently and what more we can do with the technology." And so we've taken that principle in terms of, one, making it a part of our vision at Rush and what we do and how we deploy our team. Two, we get our staff involved in that process. And then three, we talk about the outcomes. We talk about when it works. We talk about when it doesn't work. And when it does work, we also say, "Hey," You see that that didn't result in anybody losing their job. So we actually talk about that with our team so it can help ease some of that tension and have them really think through how do we evolve our workforce so they can embrace it more and more and more.

Gemma Allen

>> Staying on the topic of readiness for a second, I want to ask you both about data readiness within this deployment and this opportunity for automation? Again, we know that that has been a challenge for many industries in many spaces for many decades. How is it changing within this world of AI and automation and efficiency? And, if you're working in a billing department and you refer to something as underpaid and somebody else refers to something as, it's a different even like, taxonomy, right? That can create a whole lot of challenges in the world of structured data that's needed to feed, an automated feature like this.

Blake Evans

>> Yeah.

Gemma Allen

>> Talk me through how— what you're seeing more so and how— I'm asking you both this. How you're kind of responding to that from a solutions perspective.

Blake Evans

>> Yeah, one of the limitations— and David, I want to hear what you got to say too— but one of the limitations that we've had from a provider side is that we only know our own kind of data, our own environment. You mentioned underpayments, right? There's so much that goes into that. One is the contract that you negotiated with the payer, right? A lot of times we don't have access to that contract, but where AI and automation is really starting to help us is you can ingest different data sources so you can pull in payer policy, you can pull in your contract with the payer, you can pull in coding guidelines, right, to make your staff a lot more intelligent. So when we are trying to fight a denial or we say, hey, this is underpaid, we have the actual background in the contract and how much and where we found it in the clause. And that's the level of insight that the technology is starting to give us where we would not have had that historically.

Gemma Allen

>> And for you, David, when you're rolling out this product, Adonis products, to healthcare institutions across the US. How is that problem being managed? How front and center is this idea of structured data versus unstructured into an ingestion engine? But what do you think happened there?

David Unger

>> Yeah, it's a good question. I think, Blake, you hit on a big piece, which is data is going to be king. I don't think that's a new thing necessarily. But when you think about intelligent decisioning, whoever has the most data, the most information is going to be able to create the best insights. So data is very important. To your question about what's the process to structure it, when you think back to the adoption, 3 to 6 months, that's a big part of what's happening there is Rush is different than other health systems within its own region, certainly different than other health systems in, call it, the East Coast or the West Coast. So there certainly is still, I would say, a bit of work that needs to go in to make sure that the data fits the model. Having said that, again, there are kind of a finite number of data sources within healthcare, the EHR being the primary one. And so that's where we currently focus most of our efforts is how do we best understand the EHRs like Epic, like athenahealth, like Cerner and Oracle Health. And then moving into, as Blake mentioned, insurance policies, contract data, ERP data. So I think it's taking a structured approach to get the most important data right first and then you expand from there.

Gemma Allen

>> So moving on to the last question I have for you guys, maybe the spicier question on the pod here today. The world of AI frontier models, it's moving fast, right? We heard Claude for Healthcare is making a big push into the industry. We know ChatGPT has just released an agent plugin for Epic. It's interesting when we're moving from data to that because some folks have concerns around data retention and what that really means if you have one core orchestrator across your entire business, right? There is a dependency there for sure.

Gemma Allen

>> Mm-hmm. How—

Gemma Allen

>> and I'm going to maybe start with you, David. How do you think about that from a competitive perspective? When you see these news announcements from these big frontier models, you're like, oh God, or are you like, well, actually, yeah, give me a response.

David Unger

>> Yeah, it's— we certainly pay attention to them. I would say, though, the world of healthcare is still very, very complex. And so I think to believe that you can just install a plugin or you can take some AI features or functionality off the shelf is probably not going to be the case. Again, the way Rush and Blake operate as a health system is specific to their payer mix, to their patient population, to the demographics within that area. And so there it's not as easy as just putting an integrator in. And so as much as we believe AI and automation is going to just make things easy, there's still a lot of work that has to happen around there. So that's where we believe Adonis is fully focused within healthcare revenue cycle. It's all we do backend specifically because we believe, again, in order for AI to be useful, it has to be specific to your organization. And so I, again, we pay attention to the big names, the Epics and the ChatGPTs of the world, but we believe what we're building at Adonis is really going to move the needle and provide a lot of value for customers like Blake.

Gemma Allen

>> And Blake, I mentioned earlier this whole concept of vendor lock-in, right, which has definitely been a thing in enterprise and I'm sure in healthcare and public sector too for quite a while. We've had an interesting couple of weeks in the news. We've had all sorts of reports of, you know, there might be one orchestrator across the world, actually not just across any one enterprise and no names mentioned. But certainly it's an interesting dilemma, right? Because in some respects it's another version of, the '90s and what kind of— what about that? Talk about healthcare though. How realistic do you think that is? How realistic do you think it is that you would have one company kind of managing everything for you?

Blake Evans

>> Yeah, I doubt that that will come to fruition. Right. I think that the complexities of healthcare, not everybody can do everything well. Right. So that's the other challenge. When you look across the AI automation platform, there are thousands of companies now that are investing in AI, right? So as a healthcare provider, I have all of these options, right? I have my own EHR system, right, that can do some automation. I have these vendors that are kind of talking to me every single day about what they can do. And so I think what we have learned is how do we evaluate based off of the needs that we have? What are the problems that we're trying to solve? That's one. Two, we do short pilots, right, to really evaluate, make sure we get the full benefit of the product before we move forward in a broader capacity. And then 3, again, I'll just kind of circle back. It's important to get people involved. My frontline staff is going to have a totally different level of visibility into how they can use the automation versus my directors or myself as the top leader.

Gemma Allen

>> Okay.Yeah. Okay. So, gents, I know you're in New York from the Midwest for a couple of days. I want to talk about the next few days. What's ahead for you both professionally? What does the next— I mean, it's hard to think beyond 6 months in this world of tech right now, but Let's start with you, Blake. Well, what are the priorities between now and I guess 2027?

Blake Evans

>> Yeah. So one of our big focuses is really focusing back on the people, right? We want to really develop our workforce. We're looking at AI literacy for our team. One, how do we prepare them for the work? How do we prepare them for the future? How do we make them more analytical and problem solvers and looking at exceptions versus just looking at the mundane tasks every single day? So that's one big area of focus. And then two, We want more orchestrated AI, right? More intelligent AI versus just agentic, meaning it does a specific task. We're really trying to learn from the data and insights and become more intelligent.

Gemma Allen

>> And for you and the team at Adonis, what's ahead? You guys have a few new features coming out. You guys raising more money? Talk me through what the roadmap looks like.

David Unger

>> Yeah, so at Adonis, we recently closed our Series C just a couple of months ago. So not immediately looking to raise more money. But from a future standpoint, I agree with Blake where we're investing deeper and deeper into intelligence and decisioning. So again, how can we give our clients the best information, knowing what to work, when to work it, how to work it? And I would say the other piece we're focused on is ROI. So how do we make sure that our clients, can see and measure and understand the value of AI and automation? And that's a hard piece. Quite frankly, we talked a little bit about this earlier where ROI, when you adopt a new technology, takes a little while to figure out how do you quantify that. But I do think there's a cost of doing nothing. And so we certainly think that some of the early adopters of AI and technology, although they may not know exact ROI day one, if you fast forward 12, 18 months, I think those are going to be the organizations that are seeing major wins and successes because they took a chance and they started to do something versus wait. So those are our two core focuses. How do we do better intelligent decisioning and how do we better articulate ROI for our clients?

Gemma Allen

>> Well, that's certainly an interesting thought to leave with— the cost of doing nothing in 2026. What exactly is it? Gents, thank you so much for joining us at NYSE Wired in theCUBE.

David Unger

>> Thank you for having us.

Gemma Allen

>> I'm Gemma Allen here at theCUBE studio at the New York Stock Exchange. This is MedTech Unplugged, one of our programs at NYSE Wired. Thanks for watching.