Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Sandeep Dadlani: Why Healthcare Needs AI That Listens—Not Just Learns
149 Plays4 days ago
We don’t need more dashboards—we need more listening.
In this episode, Phil Fersht sits down with Sandeep Dadlani, the Executive Vice President and Chief Digital and Technology Officer at UnitedHealth Group, to explore what fundamental healthcare transformation looks like in the age of AI.
Sandeep dives into how UnitedHealth is deploying ambient AI and agentic systems to reduce clinical burnout, improve access to care, and, most importantly, bring empathy back into the doctor-patient relationship.
This is AI designed to serve, not replace. And it’s already in the hands of thousands of clinicians.
What You’ll Hear in 30 Minutes
- Why healthcare has a tech problem and a humanity problem
- How ambient listening AI is giving doctors their time—and their focus—back
- What “agentic AI” really means in clinical workflows
- The risks of bad AI and the urgency of building trusted systems
- How to scale AI in ways that respect patients, doctors, and outcomes
Guest Snapshots
Sandeep Dadlani is the Executive Vice President and Chief Digital & Technology Officer at UnitedHealth Group, where he leads the charge in rethinking how one of the world’s largest healthcare companies uses technology. With prior roles at Mars and Infosys, Sandeep brings a rare blend of consumer empathy and enterprise tech savvy. His current mission? Use AI to fix the system without losing sight of the people it’s meant to serve.
Timestamps
00:00 – Welcome and Introductions
01:20 – Why Healthcare Needs a Reboot
03:48 – What “Frictionless” Really Means in Care
06:35 – Agentic AI: More Than a Buzzword
08:12 – Ambient Listening: Letting Doctors Be Doctors
12:47 – Creating Space for Human Connection
14:40 – Scaling AI to 2,000+ Clinicians
16:22 – Designing Tech That Actually Helps
18:57 – Building for Real People, Not Just Processes
22:10 – Where Healthcare Transformation Goes Next
24:36 – Closing Thoughts: A More Human Future
Explore More
🌐 Learn more about UnitedHealth Group: https://www.unitedhealthgroup.com/
🔗 Follow Sandeep Dadlani on LinkedIn: https://www.linkedin.com/in/sdadlani/
🔗 Follow Phil Fersht: https://www.linkedin.com/in/pfersht/
🔗 More from HFS Research: https://www.hfsresearch.com
Recommended
Transcript
Introduction to 'From the Horse's Mouth'
00:00:11
Speaker
You're listening to From the Horse's Mouth, intrepid conversations with Phil First. Ready to meet the disruptors who are guiding us to the new great utopia by reshaping our world and pushing past corporate spin for honest conversations about the future impact of current and emerging technologies?
00:00:30
Speaker
Tune in now.
Sandeep Dadlani's Journey in Tech and Healthcare
00:00:36
Speaker
Welcome to the latest edition from the horse's mouth. And today I'm absolutely thrilled to be joined by an old friend of mine and who's a very successful voice and executive in the world of healthcare and technology these days, Sandeep Dadlani. So great to have you here, Sandeep. And great to hear a bit more about your journey and how you got to be the Chief Digital Officer for Health Group.
00:01:04
Speaker
Yeah, hey Phil, so good to be with your famous podcast, the best podcast in the world as I see it. I was thinking about you, i was on a flight and I was watching this Bob Dylan movie, A Complete Unknown, Timothee Chalamet. Have you seen it?
00:01:18
Speaker
I just sort of did that, it's fantastic. And that song there, The Times They Are are Changing, that's his sort of signature song and I thought like, Phil, you should make that your signature song because you're always saying...
00:01:31
Speaker
The times there changing, as you always say. So yeah, I think, you know, there's a little Bob Dylan in you, I think, yeah? Yeah, well, the last keynote I gave, I called it, the lines they are are blurring. Yeah, you do have a little Bob Dylan in you.
00:01:44
Speaker
No, listen, my journey is simple. I grew up in India, worked in investment banking, then worked in tech with emphasis. And then I really enjoyed this idea of trying to drive transformation and speed in large organizations.
Joining UnitedHealth Group
00:02:00
Speaker
So I joined Mars as chief digital officer and spent little over five years there and really enjoyed that journey of driving transformation in a $50 billion dollars plus company. And then UnitedHealth Group came along, really has this incredible vision, helping people live healthier lives and helping make the system work better for everyone.
00:02:19
Speaker
When I looked at the mission, the people, and just the impact at scale one could have leveraging technology, digital AI, i was like, this is the opportunity of a lifetime. So I took this role ah just under three years ago. and going to tell you the people, the purpose, the pace are all intoxicating every minute. So I'm having a ball.
00:02:45
Speaker
We've had some really good conversations about AI and healthcare and all that sort of thing. But I'm Can you tell us a little bit about how real this is in terms of, it's great to talk about it, but now we're actually really getting down and dirty with the technology.
Role of AI in Healthcare Documentation
00:03:02
Speaker
Where do you feel things are heading and what's your vision for the next sort of two or three years? Yeah, no, it's been almost fortuitous where as I joined UnitedHealth Group, this whole Chad GPT moment was happening.
00:03:18
Speaker
And I remember in December, 2022,
00:03:21
Speaker
I was in the Caribbean on vacation and and I saw TikTok of this doctor who claimed to to have submitted a large amount of documentation and research using ChatGPT.
00:03:32
Speaker
And usually it took him hours to do it. Now he'd done it in in seconds, in minutes. And he was boasting about it on TikTok. And then two days later, the same doctor showed up and said, look, I'm sorry, but ChatGPT had made up half the references and they never really existed. So I apologize for it.
00:03:51
Speaker
Of course, the first TikTok got millions of views. The second TikTok got like no views and so on. So it was a teaching moment in January of 2023 where the entire leadership of UnitedHealth Group huddled to understand this phenomenon and pull credit to the leadership.
AI's Operational Impact in Healthcare
00:04:08
Speaker
Most of the places technology is struggling to get a seat at the table. This is the place where technology is the table. And I'm going to tell you, every business leader was curious.
00:04:19
Speaker
And so I had to read up myself because I didn't know too much about it as well and and had to educate them saying, this is how chat GPT works. This is how large language model transformer models work. You fast forward about almost two and a half years.
00:04:32
Speaker
We are at this incredible moment in time where I feel that, of course, we are living with the most exciting technology of our lifetime. But most importantly, healthcare is the most exciting use case for this technology.
00:04:48
Speaker
I'd argue if I was in, I don't know, some other consumer goods company or some other sort of photo app sharing company, it wouldn't have been as mission oriented as it is here.
00:05:00
Speaker
And the thousand plus use cases that we have put into production, half of them are generative AI based, are showing real scale, real value, real impact.
00:05:11
Speaker
And now the world that we're getting into, the agentic world, in particular, I feel even more excited about the next revolution that's coming. So if I look at healthcare, the impact in these last two and a half years has largely been administrative, operational, taking away the burden from physicians, really driving far more automation and back office processes, taking costs out of the system.
00:05:39
Speaker
That journey continues arguably hundred a at 100x in this agentic world. But the next journey that unleashes is really helping access to doctors, really helping access for clinicians to really focus on the business of care, suspecting diseases early, predicting disease progression early.
00:06:05
Speaker
i mean, the ultimate goal of all this should be We can prevent predict disease and prevent disease. That's just since centuries, ah dream of mankind. So if I discount all the hype there, today we have these thousand use cases scaling across three dimensions.
00:06:24
Speaker
The consumer experience, the claim or all the administrative work, and then the care, which is really the business of care. And we're whole business, UnitedHealth Group has been...
00:06:39
Speaker
and on fire and trying to make sure these use cases scale, they make impact in a very pragmatic way, in a very responsible way. So yeah, super excited about AI and healthcare right now. So let's get into how gennii Gen AI, GenTIC can actually improve the healthcare in the United States. So we have primary care physicians who, some of them are loaded up with like a thousand patients and they can barely get through the day.
00:07:04
Speaker
i mean, I had one who He couldn't remember me. I'd see him after six months. He'd have to be completely reminded of who I am and he'd have his student doctor in doing 20 minutes of prep. And I got so fed up with it.
00:07:16
Speaker
I ended up paying for a concierge service. So I actually have a PCP now who knows who I am and I can text this guy and all this sort of stuff. So how do you feel we can see improvements in the whole system here?
00:07:29
Speaker
And what can United Health Group do to sort of drive that? Yeah, no, by the way, Phil, I don't know which doctor didn't remember you. I mean, you're a memorable guy you should be famous among all doctors.
00:07:40
Speaker
So let's take that specific experience of yours. Doctors who are seeing 15, 20, 30, 40 patients a day, spending literally maybe 5, 10 minutes per patient, then having to document all their day's work,
00:07:58
Speaker
follow-up notes, prescriptions, feed it into the EMR so that there's a record and so on. That doctor today has barely any time for himself, his family, and all clinical staff is usually overburdened.
Ambient Listening and Patient Care
00:08:15
Speaker
So To those doctors and even nurse practitioners, clinicians, right now, the healthcare industry and UnitedHealth Group is scaling out ambient listening.
00:08:25
Speaker
So in our case, 2,000 plus doctors are going to have, by the end of the year, ambient listening, which means when a doctor sees you, instead of busy typing in into a computer, which they usually do just because of compliance, they look into your eyes.
00:08:41
Speaker
They look at Phil and say, hey, Phil, how are you doing? And then they focus on the business of care. And there's, with your permission, an ambient listening scribe that's documenting in the notes accurately.
00:08:53
Speaker
And that's step one. Step two from there is the same notes is now summarized. And if the doctor approves the summary, fed into the EMR, which is an important part of the process.
00:09:06
Speaker
Step three, if we have all of Phil's records from the past, The system already knows Phil and can perhaps nudge the doctor for potentially checking for two or three conditions.
00:09:19
Speaker
Hey, Phil last time came with these and these complaints, so maybe you should check him for one, two, and three. Step four now is if the doctor's checking for those nudges and confirming,
00:09:31
Speaker
Potentially now the doctor can help even code the claim right there, as in the summary, the previous conditions, the claim. Today, providers spend 4% to 6% of their revenues in trying to get the claims coded and collected, which is ridiculous.
00:09:48
Speaker
Right now, you have about 1,300 doctors already using Ambient Scribes. By the end of the year, it'll be over 2,000. The early feedback is they have their pajama time back. they have their family time back, and so on.
00:10:01
Speaker
But wait, there's more. Now we're working with a few pieces of technologies where you could have a scenario where places which don't have primary care access at all could potentially check into primary care using an AI scribe. So if I'm a patient and I'm in like, I don't know, interior parts of some state where I don't have primary care easy access, an AI doctor can help take my information, including pictures and so on, and provide an initial diagnosis to the primary care physician who can review it, approve it. So there's a human in the loop.
00:10:37
Speaker
and can then diagnose me as well. So you're not just talking telemedicine. You're not taking it to a different level where the AI is capturing information. You see where all this is going.
00:10:49
Speaker
This is going to a place where we have primary care access for all, doctors, nurses, and clinicians focus on the care part of healthcare. care The care part of healthcare, care that's what they're supposed to do.
00:11:04
Speaker
That's what they dreamt of doing. they They grew up saying, I will care for people like Phil personally. And they don't have to deal with administrivia. Everybody can schedule appointments easily and so on and so forth. So that particular example you gave is a good example of the clear three consumer claim care.
00:11:22
Speaker
That whole care part, We just went through it beautifully in terms of what could happen. And we are well on our way. We are a player that can make a massive impact on the system.
00:11:33
Speaker
We now partner with ambient scribing companies to take the same capabilities to other healthcare companies. So it's one thing saying, oh, UnitedHealth Group, what are you doing? Yes. But remember, have the privilege of helping others as well.
00:11:47
Speaker
It doesn't work if I just rocket up my AI and nobody else gets it. No, everybody else can help get their capabilities up and we're here to help actually.
Balancing Innovation and Regulation
00:11:57
Speaker
So how do you balance this innovation like with this whole regulation of healthcare and the current environment we're in?
00:12:05
Speaker
Yeah, great question, Phil. And look, I mean, you talk to so many corporate executives and everyone's playing a little bit of a balancing act. I'd say in healthcare, it's a higher responsibility.
00:12:17
Speaker
A, it's very regulated. You're right. It is regulated. And the regulation is there for a reason. Because individual players should not be allowed to play around with serious business of healthcare.
00:12:28
Speaker
The regulation over the years has... in some ways helped, in some ways it has not helped. It has created more sort of bureaucracy and so and so forth. And so how do you foster innovation in that scenario?
00:12:41
Speaker
What we've done, but particularly with ai is first made sure we armed people with AI and trained them on AI.
AI Upskilling Initiatives
00:12:50
Speaker
So today, and just in my tech shop, 23,000 engineers code using GitHub Copilot X. And until date, we have accepted almost 100 million lines of code. I say, you know, Satya and Microsoft might be proud of that number because there not too many large non-big tech organizations who do so much of wipe coding.
00:13:11
Speaker
We've accepted close to 100 million lines of code, which was suggested by an AI. Our acceptance rate is about 18%. So our engineers accept suggestions 18% of the time and 82% of the time they're like, no, I think this needs changes.
00:13:28
Speaker
That's just the engineers. who are using AI to code. Now let's get further. I have about 13 and a half thousand engineers who have signed up to become LLM agent masters by the end of the year.
00:13:46
Speaker
I already have on my United AI studio, which is a single place to build AI, to fine tune LLMs, to build agents. I have an average of two and a half thousand engineers that access it every month.
00:13:59
Speaker
So we've armed everybody with AI tools. Then we start bubbling up use cases. So there's a groundswell of use cases that filter up every month.
00:14:10
Speaker
Every business reviews the use cases. So every business CEO reviews the use cases. Top down, all business CEOs are in AI immersion sessions every month. demos, out of the possible sessions. So this is coming from the leadership top down.
00:14:27
Speaker
And we prioritize the use cases based on which of the top 20 that are adding value, that are making patient impact, physician impact, that are scaling, that are profitable, and so and so forth.
00:14:39
Speaker
And the signature event that brings all this together is something called Tech Tank. And I showed it in your last shindig in in New York City, where it's almost like Shark Tank, except it's live webcast to all the employees and the top six use cases present.
00:14:57
Speaker
And you know there's a little bit of play and fun and there's prizes, but it's signaling to an organization that this is central to our future.
00:15:07
Speaker
This is not the other thing we do. Or let's do the other AI thing that can be taken to the board for a shiny demo. This is a formal board agenda. This is a monthly agenda for all CEOs and our group CEO.
00:15:22
Speaker
And there are tens of thousands of engineers working on it. Our ah strategic vendor partners, since you work a lot with SIs, They have strict instructions to only have engineers that are already trained on AI on the account.
00:15:36
Speaker
There should be nobody else. Why would any SI put a non-AI trained person on the ground, period? So I think it changes the business model, the delivery mechanism, the top-down prioritization, the bottoms-up innovation.
00:15:50
Speaker
And it's fun to see this throbbing system create more value in healthcare. So you talk about vibe coding and trying to sort of rebalance how tech people operate.
00:16:03
Speaker
We had a good conversation about this recently. Companies have to really lean into their talent more than ever, right? How do you do it? How are you working with people who've been programming one way, working one way for all their careers, and now they need to think differently and stuff? What are the big challenges that you're experiencing, Sandeep?
00:16:22
Speaker
Look, this is a reasonably big tech shop. So I have tens of thousands of engineers internally and many thousands of engineers as contractors or vendors to our system.
00:16:35
Speaker
And we're all doing important work. Each one is building a system that is going to impact thousands and millions of lives. So this is not for the faint hearted.
AI Enhancing Coding Productivity
00:16:42
Speaker
This is not some random back office system that nobody will see.
00:16:46
Speaker
So first we created this AI Dojo upskilling class. We worked with our partners. We created this amazing upskilling class, which is about 60 hours of work.
00:16:57
Speaker
And in 60 hours, a Python programmer can become an LLM tuning agent building machine. And they have to complete a capstone real project to prove that and clear that.
00:17:11
Speaker
And we mandated that by the end of this year, everyone has to take this. Period. That's the mandate. Everyone has to take it. If, you know, we want to be thriving in this healthcare tech ecosystem. So that's point number one.
00:17:25
Speaker
Point number two is we've had several examples of success. Like yesterday, I was in a review and somebody just did an Angular upgrade, like, you know, a front-end upgrade to React or Angular from like an old front-end.
00:17:39
Speaker
And normally we say, oh, yeah, it usually is 20% more productive to use AI. This guy took a six-week project and finished it in minutes. In minutes, because you can ask a LLM to now do it for you and then check the code and then check the output.
00:17:54
Speaker
And the unit testing was using an LLM. And the test cases were using an LLM. And he did it in minutes. Well, every week we see examples like that and celebrate the bedouzles out of them. Like we really celebrate them. It's on the intranet.
00:18:11
Speaker
It's in the town hall. we make heroes out of these wonderful, wonderful employees who decided to be bolder. And so it's a cultural celebration of the best, the mandate on training, the upskilling.
00:18:26
Speaker
Hiring from the outside into ah tech is only AI. If you don't have ai actual AI projects or actual AI skills on your resume, you can't get in.
00:18:37
Speaker
In fact, it's funny, I saw your LinkedIn post about freshers. Yeah. I don't know how you feel about it, but there's different philosophies. or Oh, in this world, some people say, oh, freshers will be not be useful. And some people say freshers will be useful. So actually, I'd love your thoughts again on that. What do you think now?
00:18:54
Speaker
Will freshers be useful or not? Yes, I think that... We're getting so deep in using AI to do our jobs for us us that we run the risk of losing our identities if we're not careful.
00:19:08
Speaker
It's like I'm a writer and if I put something I wrote through ChatGPT and ask it to rewrite it for me, I feel really dirty and it's not real. I have to be more authentic. I have to be more authentic.
00:19:19
Speaker
and The more I evolve, the more I'm using these tools differently. And I think when we're talking about bringing the bright kids out of college, I think you've got an opportunity to mold young minds the way you want to mold them, train them, develop them, have them operate you know without all the baggage from the past. You almost want people without baggage.
00:19:42
Speaker
And we talk about our lovely services industry. These are companies with hundreds of thousands of people. And you want to be different and distinctive, then become like a university yourself.
00:19:54
Speaker
Become a place where people get excited to come to work. You build a culture. You've got to have people want to work with you as a business. And I really believe if you just have a very Neolithic view of AI and just think about how can we just get rid of people,
00:20:12
Speaker
How do we just save money and stuff like that? You're going to lose your identity very quickly as a business. And I worry when I see in our industry, for example, clients go, oh, we want 20% cost reduction because you use AI now.
00:20:27
Speaker
Great. So how are you going to do that? Are we just going to fire people? are you actually going to really figure out How can we do things more intelligently, faster, slicker, better, and have a culture of success so we can focus on doing higher value things for our clients?
00:20:44
Speaker
So that means our people need to think differently. You're starting to think differently. And that's where this is going. And my advice is you have to lean into your people more than ever if you want to lean into AI technology.
00:20:57
Speaker
If you're just obsessed about the tech and removing people, you'll kill your culture. You'll become like a depressing place to be. And I've looked at the data. 45% of employees right now are scared of AI.
00:21:11
Speaker
They're worried about their jobs. A third are sort of uncertain and only about 15 to 20% are really embracing us. So the majority of the workforce right now is living in particularly in white collar jobs are living in fear.
00:21:25
Speaker
And we need to get rid of that fear. and So well said. And you know I think from my perspective, just echoing with some examples, I hang with our intern batch quite a bit.
00:21:37
Speaker
And what I'm finding, Phil, is that the interns who come in who are like in the third year of undergrad school or or the final year, they're waiting to graduate. They're dreaming of a different future. Their questions are not trying to build a horseless carriage.
00:21:55
Speaker
So if I take a middle manager and say, oh, do you know AI? Yes, I know AI. Okay, cool. How do you use it? Well, I do wipe coding. Okay, how do you wipe code? Well, this is what I do. I'm trying to take my SDLC process and I am trying to automate it by 20%.
00:22:10
Speaker
to build this app. Then when I go ahead with the interns and say, hi, how would you build this app? They're like, well, we just prompt the LLM to build the app. And, oh, by the way, the app front end will not be the front end.
00:22:24
Speaker
Why are you building anything in Angular or React? This will just be an LLM. So why are you even building a front end to your app again? And what wait, wait. So someone in the system, and i mean, there are obviously great exceptions, but someone in the system with AI is building a horseless carriage.
00:22:41
Speaker
Someone new to the system, out of school, has started thinking, I'm building a next generation automobile. And so honestly, we've actually increased our hiring of freshers. I am amazed at anyone who's saying freshers will be less useful.
00:23:00
Speaker
Are you kidding me? They're coming in, to your point, they are already switched on to a different world. It's like the iPhone. Your kids took to it faster than you did and so on. so So I feel it's a great time to graduate with the right skills and to build next generation automobiles.
00:23:19
Speaker
And people like me need to really worry about my skills and not build horseless carriages and actually have to just reinvent, like completely reinvent the experience.
Generative AI in Patient Experience
00:23:28
Speaker
So one of the things in a consumer track,
00:23:31
Speaker
is how we're imagining the future, Phil, is I, as a patient, don't need to interact with, you know, a website or an app. I have my own personal assistant that knows me, that knows my health, that knows my secret diagnosis, which nobody else should know, that knows my language, my tone, my swear words, anything, right?
00:23:53
Speaker
And it's talking to me. It has my health history, which no single healthcare institution has. And it navigates me through the healthcare system. It schedules my appointments, my checkups, my blood tests. It interprets my blood tests.
00:24:06
Speaker
It takes me to the radiology center. It just helps me navigate this extremely complex system. We're launching our first generative AI assistant this quarter, which I actually announced at your summit.
00:24:22
Speaker
on the UHC side, then we're going to launch a similar system on the Optum side before the end of the year. And then we start connecting the experiences so I can do my ah RX refills or schedule a doctor or check my claim or get my member ID or get my benefits card or see my labs, all of that in one connected experience.
00:24:45
Speaker
Now you start doing it with other healthcare systems. You start building your, you know, the vision of the whole consumer space. Just initial pilots, the call center calls have gone down.
00:24:58
Speaker
it's incredible just seeing all that digital activation happening. So yeah, I need more freshers with native AI skills who don't build horseless carriages. And it's the opportunity to fix healthcare. That's the only way.
00:25:11
Speaker
I think it's fascinating that we're almost looking at bringing people into the companies who can totally think differently and immerse themselves and in the business. I mean, we're almost looking at creating entirely different jobs that didn't exist before.
00:25:27
Speaker
And, you know, I take the attitude, for example, with services, We were an industry that thrived on skills at scale tied to common technology platforms. Now we're looking at scale again, but tied to AI platforms.
00:25:42
Speaker
So the principles are the same. It's the technology that's different and it's the skills that we need are different. And we're right in the middle of, we're not even in the middle, we're at the beginning of a massive transition to a very different world.
00:25:57
Speaker
I mean, do you think we're going to make it there or do you think that there's going to be a lot more roadkill along the way as we, because this is a big societal change as well as a business change, right? so Well, I'm an optimist on this wave of change. So I've gone through my set of dissolutions on, i don't know, blockchain or you have all kinds of fads that came and went. And I'm guilty of even celebrating the NFTs, which were launched like and a little whole crypto NFTs.
00:26:29
Speaker
But I think this time is different. Even if I take away 90% of the hype out there, and there probably is 90% of the hype out there. So if I take 10% of the capabilities as true, I'm already seeing pragmatic,
00:26:46
Speaker
results at scale. And I'm sitting on a thousand use cases.
Democratizing Technology Through AI
00:26:53
Speaker
I think there are five or six being generated every week that are new, that some will pilot, some will scale, some will fail.
00:27:00
Speaker
I'm seeing an entire organization, a Fortune 3 organization, reacting, learning at scale. And I'm seeing magical productivity gains. So I think this time is truly different.
00:27:13
Speaker
I'm seeing talent being transformed at scale. Here's an interesting experiment we're trying with Microsoft. So they introduced Copilot Innovate Studio, this agentic studio where you can build your own agents.
00:27:26
Speaker
And by the way, everybody has one and so and so forth. So we gave this studio to about 1,500 tech people and 1,500 ops people. So the first time we've taken non-tech people, this is special.
00:27:39
Speaker
You know, tech people can do what they want, build agents. Well, non-tech people as well. So we've launched that experiment last month and in a controlled environment, this is, you know, we've protected it from anything in production. This is a test environment.
00:27:53
Speaker
But we want to see in six months what they produce. It's a social experiment because Yes, there are lots of podcasts where people say, oh yeah, this thing can speak English, so now everybody can code. Well, not really, really. You've got commission servers, you've got a test, you've got a cyber secure it, all that stuff is.
00:28:08
Speaker
So this is a social experiment at scale, 3,000 people. I'll come back and report to you in November in your next podcast saying what happened. Did operational aid people produce more interesting use cases?
00:28:23
Speaker
Did they make it safe? Did they make it compliant? I don't know. If they do get success, you're now liberating 95% of your organization that never touched tech to building tech.
00:28:37
Speaker
And guess what happens then? i mean, it's the force of the org is unleashed to build whatever tool you can. Now, there are compliance, cybersecurity, responsible AI, which we are a big, big believer in. And we really sort of govern our responsible AI framework.
00:28:54
Speaker
But I think we're in a moment where technology is truly getting democratized, empowered. I used to put this 100x thing in my previous jobs as well. I think we're getting to 100x.
00:29:06
Speaker
Wow. Absolutely right. And um as you look, at your sort of talent acquisition strategy now, i mean, what are you looking for in terms of like, is it particular qualifications? Is it a different type of personality trait? I mean, what is it that you're prioritizing, Sandeep?
Importance of Learnability in Talent Acquisition
00:29:26
Speaker
Well, one of my first companies taught me this idea of learnability, that if you can learn anything, you can do everything. And so the first thing, least I look for, whether a fresh graduate or it's a middle manager or it's a senior leader, is how much have you learned last year?
00:29:45
Speaker
Of course, you learned something 20 years back, but what skills you And that has to apply to me as well. Have I learned something new last year? So learnability is huge. Look, real projects involve real people, challenges, failure.
00:30:00
Speaker
I like people who... have real stories to tell. We've grappled with change, who've had hand-to-hand combat on trying to convert somebody else into new technologies.
00:30:12
Speaker
If people walk in and say, hey, oh my God, I had these three fantastic projects and they went through smoothly and I saved hundreds of millions of dollars or I generated billion dollars of revenues. Yeah, that doesn't sound right.
00:30:23
Speaker
That's not real. Real stories are scars. I like people with scars, people who have worked with other real people. So in this age of ai in particular, we can easily get seduced by the technology.
00:30:38
Speaker
It does not change a few important things. Are you zero distance to the problem? Something I learned in a previous company as well. Are you coding and solving these problems in a conference room in Minnesota or are you in the clinics, in the waiting room, watching how nurses and patients struggle?
00:30:55
Speaker
Are you in the call center taking calls? That's the one of the first things I did. So about one day a month, we do this. We're either in a call center taking calls or in clinics and we just get real about problems.
00:31:07
Speaker
And then do you deliver every four weeks? So real experience, learnability, zero distance experiences, curiosity. These things are just, you can't compromise these things.
00:31:18
Speaker
You can have all the AI you can learn, but but these things can supercharge you to the next level. Supercharge you to the next level. I um will take that advice. i think that's fantastic.
00:31:30
Speaker
And learnability, I mean, we used to talk about this, but it's so true now of that first question I think I'm going to ask someone in my next interview interview is what have you learned in the last year?
00:31:41
Speaker
Just tell me what you've learned. Because I'm telling you, I work with people and I don't think they're learning a lot of people. a lot of people are, but some people, they do things the same way. They've always done it. It's like going to a dentist and they're using equipment from 20 years ago and they haven't like kept up with the latest techniques or ideas, that
AI in Research and Operations Efficiency
00:32:00
Speaker
sort of thing. like We should all be like that.
00:32:02
Speaker
We should all be thinking, how can i be a better analyst? Or I'm just going to do things the same way I've been doing all my career. I mean, just simple tools, right? Deep research through any of these tools, ChatGPT Grok or Gemini or others.
00:32:17
Speaker
I don't think people realize how much work it can save them, including consulting work. Like the people who are still commissioning consultants to do some research on some project.
00:32:29
Speaker
I was in a meeting where we were discussing media allotment to different channels for certain outcomes. And I just asked Deep Research that question. And it gave me the precise research on if you want these outcomes, then here should be the channel allotment for your media.
00:32:44
Speaker
Think about that. Like, and we're a healthcare company. We don't do much media like this. Entire media companies is running on this. So that's a skill. And those who learn it, you know, I like pop songs and metaphors. So the song that plays every time somebody shows off their new superpower is Imagine Dragons' Thunder.
00:33:03
Speaker
I was the lightning or the thunder. You know that song? You like that song? Yeah. It's all right. You get a new superpower, Phil. Promise me, you'll start singing in your head, I'm the lightning before the thunder.
Wrap-up and Future of AI in Healthcare
00:33:14
Speaker
Because think this is that kind of moment.
00:33:17
Speaker
This is that kind of moment. We are beginning to show off new skills in the moment, in day-to-day life, in our enterprise work. We are the lightning before the thunder.
00:33:29
Speaker
Do you know what? I think yeah I'm going to go listen to it now. and We should probably play these tunes to fade out our podcast today because this has been a fantastic conversation, Sandy. And I can't wait to catch up with you again in a few short months to see how the concierge service that you're building, the Gentip concierge, what's the official name for your solution?
00:33:55
Speaker
Ask Ask AI. Ask AI. Yeah. Okay. Yeah, yeah. Ask AI. Employees have it. I already chat with it, you know, and in the app already got 111 million logins in this Q2 in the second quarter.
00:34:11
Speaker
Imagine with this feature on, it's number one, number two healthcare app in the world. It's just like how much logins, how much engagement it will get. So we're super excited of creating that whole navigation system.
00:34:22
Speaker
Ask AI. You heard it first. Thank you, Sandy. Till the next time. Thank you, Phil. Remember, you're the lightning before the thunder.
00:34:33
Speaker
Thanks for tuning in to From the Horse's Mouth, intrepid conversations with Phil First. Remember to follow Phil on LinkedIn and subscribe and like on YouTube, Apple Podcasts, Spotify, or your favorite platform for no-nonsense takes on the intricate dance between technology, business, and ideological systems.
00:34:53
Speaker
Got something to add to the discussion? Let's have it. Drop us a line at fromthehorsesmouth at hfsresearch.com or connect with Phil on LinkedIn.