Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
Welcome to Tech It To The Limit, the humorous and surprisingly informative podcast that
(00:22):
makes digital innovation and healthcare as entertaining as it is relevant.
I'm Sarah Harper.
And I'm Elliot Wilson, and we're here to pull back the curtain on the world of digital
transformation in healthcare.
Don't worry, you don't need a medical degree to join in on the fun, just a sense of humor
and a penchant for all things health tech.
So buckle up folks, it's time to Tech It To The Limit.
(00:45):
What's up, Elliot?
What's up, Sarah?
How are you?
I'm excited to be here with you on episode eight.
I'm so pumped.
We are almost at the end of the season.
I'm actually like, it's a little bittersweet moment for me because I love Tech It To The
(01:06):
Limit.
And when we go dormant, like the bears in the Northwoods, I miss you.
I miss you too.
But you know, I get to, we snack on our little hibernation nuggets to keep us through the
winter, right?
To keep us warm.
I don't know what I'm saying.
I miss you too, Sarah, through the winter months, but it means that I have something
to look forward to.
Why don't you tell me about health?
(01:28):
Oh my gosh, health was amazing.
So you know, the energy was really palpable.
And I like going to health because it has a certain vibe to it, which is this one of
positivity where venture capital is coming together with founders.
And there's just all these founders all over the place that just want to change the world.
(01:49):
And little do they know that they're going to get swallowed up by either their competitors
or, you know, the lack of funding.
And that and like the high profile guests that they have, it's just, it's a lot of fun
to go.
And they had some serious speakers there this year.
They had, you know, some celebrities, Lenny Kravitz or Lance Bass, Halle Berry came on
(02:10):
and was launching.
Please tell me what these people have to do with healthcare technology.
I'm curious.
Yeah.
I'm not sure about Lenny Kravitz.
I think he was just there.
I saw him on stage.
I think he was just there to look really good.
And let me tell you, he looks really good.
That's his job, isn't it?
I know.
I kind of think it is.
But Lance Bass has has diabetes.
(02:31):
And so he's done a lot of work in the diabetic space.
And Halle Berry was there and was launching a product line called Respin, which is all
about women's health.
And so it's yeah, it was it was really good.
And she was a very dynamic speaker.
But between the celebrities, they also had politicians and administrators like Chelsea
(02:53):
Clinton was there.
She was wonderful to listen to.
I didn't get to see Jill Biden.
I was already gone before they came.
But apparently like they really deeped up to security as one would expect.
Seema Verma was there.
There was not as much security at the beef look for you though.
No, no, no, no, you know, kind of people like that.
Then they had provider leaders, you know, they had CEOs from Cleveland Clinic, Kaiser,
(03:16):
Health Plans, etc.
And then some of the big tech AI heavy hitters like Eric Horvitz, the chief science officer
at Microsoft or Kim Powell, the VP of healthcare for NVIDIA talking about AI.
Greg Carrotto from the Google head of health AI, like they had some big heavy hitters there
(03:37):
in that space.
I didn't do a lot of listening to the sessions.
That's not really what I was there for the ones that I did.
I didn't hear a lot of new stuff.
But what I love to go there is just looking for new technology to see what out there.
And they had a lot of different pavilions all around this year.
And they really threw their money into this pavilion kind of set up for their conference.
(03:58):
And so there was country based pavilions like Korea, Israel, there was a global partners,
pavilion from countries all over the place.
The digital health help foundation had an amazing pavilion and I was a judge for their
awards.
So that was kind of fun to see them there.
Thanks.
Yeah.
It sounds like the world's fair for Hal's tuck.
(04:19):
Oh my God, it totally was.
I think like exactly what it was.
They had food as medicine and oral health, which were two new topics.
I shouldn't say new topics.
They're not new topics, but they were kind of new to this space.
That was really kind of cool.
But my favorite pavilion was the nurse innovators pavilion.
So I got to meet Rebecca Love who was helping to put on that pavilion there.
(04:42):
And it was just all new tech or new startups by nurse innovators.
And that was really neat to see.
So there was this huge focus on nurses this year.
Oh, I just said, sweet.
I was going to ask, did you see Bonnie Clipper there?
I did.
No, we totally did.
We caught up with each other and said, hi.
(05:03):
She is hysterical as per usual.
And then we caught up again at the Busta Rhymes concert, which by the way, that's the other
thing that was there.
What?
Yeah.
So Busta Rhymes was like the industry night star.
So it was a Busta Rhymes concert.
This is a giant cash bath of a conference.
That's my goats.
Way too much money.
(05:24):
Way too much money is kind of out there.
The Busta Rhymes thing was kind of lost on me.
I mean, his heyday, like late 90s, I was a chubby closeted white kid in musical theater.
So like I wasn't doing a lot of listening to Busta Rhymes at that time of my life.
To be fair, you're still a white kid.
(05:44):
That is very fair.
Very fair.
But at least I'm not closeted anymore.
So I've got that going for me.
But I will sum it up.
There was this great quote I did listen to that I heard from Matthew Lungren, the chief
data science officer at Microsoft that I really liked because there's all this buzz about
AI and will it replace doctors and nurses and other clinicians.
(06:09):
And he said, AI will save time before it saves lives.
And I love that.
Isn't that a great quote?
Like, yes, that is where we need to be focusing AI, taking out, adding time back into our clinicians
days to provide better care.
(06:29):
Hashtag machines make care more human, right?
Yeah.
We keep coming back to that, man.
This is all your genius, changing the world.
One hashtag at a time.
But enough about my hashtags.
You were busy this month too.
Yeah.
Actually, I was co-directing the Rural Health Symposium, which is our flagship CME course
(06:52):
for our Mayo Clinic Health System where I work.
And we had some, I had an amazing time giving out free hugs, connecting with people from
around the world who wanted to think about what types of cross ecosystem and even cross
industry partnerships are needed in health care so that we can truly transform at scale.
(07:16):
So we had Santo Cruz, who's the Chief Legal Officer of CentraCare, which is a Midwest-based
health system, alongside Tina May from Land O'Lakes talking about the American Connection
Project.
And one of my favorite nuggets from their panel was Santo Cruz was quoting the Rural
Movers Study by Ben Winchester, who's a researcher at University of Minnesota, kind of shedding
(07:42):
some light, busting some myths really, that like there's this brain drain from rural America
into the cities.
He was really using an evidence-based approach to show that there's actually a brain gain
in rural America and that the median age of folks who are moving back is 36.
And the top reasons for their going have nothing to do with the economy or health care.
(08:05):
It's all about the lifestyle, which I found so interesting as we think about how to make
rural health care viable in the long term.
And we try to recruit a workforce that is sustainable in those cities.
We need to think about the whole community and not just like what kind of benefits packages
are we offering, right?
(08:25):
And so one of my favorite moments of the symposium is we actually highlighted some local artists
in the evening at our reception, not Lenny Kravitz, like the Lenny of Rochester, Minnesota,
and otherwise known as Clay Fulton, who's one of my favorite local musicians and songwriters.
And then Samuel Hawkins, who's a spoken word artist, kicked off the evening with some of
(08:48):
his poems that are inspired by community.
And at first, a lot of people were like, what is this?
I don't really get the connection.
It was perfect kind of tying this message of artists build a sense of community and
almost like a hearth that you can gather around in my town.
And that's really what our leaders need to be thinking about is they think about investing
(09:11):
in the future of health care.
Another favorite moment was our AI panel, probably not as expensive as the one at health, but
it featured Eric Karnesh of Mayo Clinic Platform, my medical director from the community practice
Dr. Rajiv Shadri, and also Dr. Matthew Kalsram from our generative AI program here at Mayo
Clinic.
(09:32):
And I thought that they had some wonderful insights to share.
Eric talking about design in health care technology and like how we need to make it seamless.
I mean, I know we've been having this conversation for decades, but he asked the question, did
anyone need to be taught how to use Amazon Prime?
Like, come on, people, right?
(09:53):
We can do this.
What a great point.
What a great point.
Yeah.
And then kind of dovetailing with what you mentioned earlier from Matthew Lundgram's quote about
AI saving time before it saved lives.
Dr. Shadri made the case that the only way for community and rural health systems to truly
survive and thrive is through the appropriate use of AI and automation at this moment of
(10:18):
time.
And I think what he was getting at was, do we need to save people time so that they actually
enjoy working in health care and stay, right?
And then my final quote that I'll share was from Dr. Matthew Caltram when he was saying
that the greatest opportunity for AI is to bring humanity back into the conversation
(10:39):
between doctors and patients.
So again, hashtag machines make healthcare more human, right?
Yeah, I'm right.
So an amazing course.
Definitely a highlight of my year and excited to be offering that again next year.
So cool.
So cool.
Sounds like an amazing program.
I wish I could have been there to get a free hug from you as well as all of these amazing
(11:03):
nuggets of wisdom and learning.
I am hugging you through the screen right now.
Well, speaking of making machines more human.
Who are we hosting on the show?
Our guest today, this is so exciting, really big get.
We are interviewing chat GPT here on the podcast today.
(11:23):
We thought it would be really fun and interesting if we trained a GPT or chat GPT and prompted
it to be a digital health expert.
And so we did and we are hosting and interviewing him this afternoon.
(11:45):
Yes, very excited to have Kat I farted on the show.
For those of you who don't know that Shaday Pette actually means Kat I farted in French,
you can check out the link to the video that we're going to put in the show notes.
Yes.
Just for a little, you know, who's the giggles for those folks that may not know what chat
(12:07):
GPT is or what a GPT is and at all.
So it might be good to take two minutes and kind of explain what the heck we're talking
about here.
So in its simplest terms, chat GPT is a chat bot that's powered by artificial intelligence,
but it's not like a clunky one word answer bot of the old days.
(12:29):
No, chat GPT is a large language model, which means it's trained on oodles and oodles of
text mostly scraped from Reddit to understand, generate and mimic human like conversation.
So imagine a digital parent, but a really, really smart one that understands context
(12:51):
and nuance.
So how does it work?
So it's built on a family of AI models called GPT or generative pre trained transformer.
This one happens to be developed by open AI.
Wait, are we talking about shorts and negative transformer?
Like is this?
So no, do you mean a terminator?
(13:14):
I do.
Sorry, I couldn't.
No, this is a different robot in disguise.
Anyway, so each word in GPT, it kind of reveals a little bit about how this all works.
So generative, like this model doesn't just repeat answers, it's memorized, it generates
responses based on a probability, kind of like predictive text, but like on caffeine.
(13:38):
So when you ask a question, it crafts a reply.
That's a new one every time.
Pre trained means that this model isn't just built from scratch every time you use it or
play with it.
You know, it's pre trained on this huge data set, books, articles, web pages, scraped from
Reddit.
(13:58):
So it already has this baseline understanding of language or at least the crude language
they use on Reddit.
And then finally, a transformer.
So this is the architecture that makes GPT so good at handling language.
It breaks down your input and weighs words against each other and in relation to each
other, like figuring out if bank means a river bank or some place that you put your money
(14:24):
or other deposits.
Like the kind of deposits my dog leaves at the park?
Sure.
Sure.
That's right.
That's true.
Okay.
So that's what we mean by chat GPT.
That's what a GPT does.
And that's why we thought it would be fun to play with one today.
(14:45):
It is going to be tremendously fun.
And I will say, Elliot, that was an exceptional description of what chat GPT is.
But if you're listening to this show and you don't know what chat GPT is, I don't know
what to say.
It means that you have a curious mind and you love hearing about healthcare and thinking
(15:07):
about it from a digital perspective.
And we're so glad to have you here on our show, even if you don't know what chat GPT
is.
All right.
You're right.
You're right.
I should channel my inner compassion.
Yeah.
Yes.
Anyway, go.
Let's move to the game.
Oh, I'm so excited about this new game.
So when we come back, a new game.
A new game.
(15:28):
Turkey or tech.
Welcome back listeners.
Who's ready for some awkward holiday?
Get together.
Me, me, I am.
(15:50):
I can't wait.
Said no American millennial ever.
No, seriously.
I actually love Thanksgiving.
I mean, it's not just because of the gravy and the stuffing, but that has a lot to do
with it.
I love seeing my grandmother.
I love getting together with my niece and nephew.
And you know, just like doing the gathering thing, doing the nesting, the warmth, drinking
(16:15):
the red wine, you know, all the, all the great pieces of it.
But I digress.
In the spirit of this, our most overindulgent annual gathering in America.
This game is going to pitch real healthcare technology innovations up against some imaginary
(16:35):
ones, all united in the name of solving Thanksgiving dinner induced health woes, be they physical
or mental in nature.
Okay.
What are the rules of this Turkey or tech game?
Your tech.
So each of us is going to take turns reading a product.
(16:56):
And the other host, as you may have guessed it, is going to shout Turkey if they think
it's fake product or tech if they think it's a real product.
The loser of the game has to take a selfie dressed as a Turkey and post it on LinkedIn
with the caption, I'm a Turkey on Thanksgiving day.
(17:18):
Yes.
This sounds just perfect.
I can't wait to see what you look like with a little gobbler.
I mean, I already sent a Turkey inflatable costume to your home address from prime.
So it's, you'll be ready when you, when the day comes.
(17:40):
Let's start out with like best of five.
Do you want to do that and see, see where that gets us?
Best of five.
Yes.
Sweet.
I love it.
Would you like to go first, Mr. Wilson?
Go first.
As the reader or as the guesser?
You decide.
You know, you're in charge.
Okay.
I'm going to give, I'm going to read first.
Okay, sweet.
(18:00):
Okay.
Are you ready?
Ready.
Hit me.
The calm patch.
This is a patch placed on the back of your neck that's claimed to induce relaxation post
meal, promising digestion as smooth as a spa day.
(18:20):
Okay.
That's not why I thought I would need the calm patch at my Thanksgiving dinner, but I'm
going to go in and say Turkey.
That just sounds ridiculous.
That's true.
No, it's, I mean, that's correct.
That's correct.
Okay.
True facts.
All right.
One point.
(18:41):
Gobble, gobble, gobble.
It said a ding, ding, ding for Sarah.
All right.
Are you ready?
Hit me.
All right.
Who is your personal accountability buddy for Thanksgiving recovery?
This wearable tracks everything from strain.
Yep.
Both from workouts and family stress to sleep.
Oh, I thought I was from pooping.
Did you stay up too late roasting the turkey?
(19:02):
Whoops.
Got the data.
Did you walk off a bit of stuffing?
Whoops.
Analyzing that strain score, reminding you that even the walk to the fridge counts.
Thanksgiving, you'll have enough metrics to justify that extra slice of pie or to motivate
yourself to take on leftovers aerobically.
The best part, it calculates a recovery score, which whoop fans swear by is the ultimate
(19:27):
excuse to keep lying down or when it's high, we'll call to go hit them all for some post
holiday steps.
I, I got to go Turkey.
I got to go cold turkey on that.
It's a jack.
Whoop.com, baby.
(19:48):
No.
No.
I thought for sure.
I'm sorry.
Because it just seemed, you know, it.
Yeah.
I had to mask it, you know, a little bit, a little bit.
I had to mask it with my humor.
All right.
Sharpers up.
One nil.
All right.
(20:08):
Next up is Quell.
Well, is a wearable device that uses nerve stimulation to help relieve chronic pain.
It's like having a secret weapon against chronic pain strapped right to your leg, delivering
nerve stimulation to keep those aches at bay for those marathon cooks all day long in the
kitchen, tireless chatters and anyone who's ever said my back needs a vacation.
(20:33):
Oh my God, like all of my uncles.
Right.
So go ahead.
Well, well, well, well, I mean, it has sort of a Game of Thrones type of like sound to
the name.
I'm going to go with.
I'm going to go with tech.
(20:56):
It is tech.
Sorry.
I was a bit loud.
Listeners.
I was just testing my mic.
I haven't won a game all season.
All right.
I want this one.
All right.
Two, no, two, no.
You ready?
All right.
This is upright to go.
(21:17):
No, sorry.
Upright go to the upright go to posture trainer is your secret weapon against slouching at
the holiday table because nothing says I'm grateful quite like a good sure this sweet
gadget attaches to your upper back and gives a little vibration.
If you start hunching over that mashed potato mountain, perfect for the relative.
(21:40):
Those who comment on your posture straight upright go to takes over as your personal posture
police buzzing you back to attention with just the right mix of encouragement and exasperation.
What's exasperation have to do with it?
Whatever.
Thanks chat.
Dbt Thanksgiving bonus.
It syncs with an app to give you a posture score so you can complete compete for the
(22:01):
title of most polite guest at the table or get passive aggressively buzzed throughout
the entire meal.
So I own an upright go.
So it's tech.
Okay.
You could have just told me from the get go, but I guess our listeners.
Ooh, no pun intended.
I'm the get go.
(22:23):
All right.
The score is two to one.
Your favor.
I'm getting a little bit nervous about wearing a turkey costume.
Okay.
So are you ready?
I'm ready.
Tell me.
So for tech.
Introducing consistency sense.
The ultimate Thanksgiving sous chef for your sauces and soups with a touch of sauce and
(22:48):
a dash of genius.
This smart utensil is here to make sure you don't serve lumpy gravy or watery soup.
So forget the guesswork and say goodbye to endless stirring and second guessing.
So here's how it works.
You dip the consistency sense in your pot and it gets to work measuring texture and
(23:08):
thickness with scientific precision.
And when your gravy hits that perfect, rich, smooth consistency, it buzzes to let you know
it's ready for the table.
But here's the kicker.
It doesn't just tell you when it's right.
It tells you right on time.
So there's no more scrambling to fix a chunky sauce before the turkey hits the table.
And you'll have everybody saying, you made this from scratch.
(23:31):
Okay, number one, this has a thing to do with health care.
Except for maybe giving everybody coronary artery disease.
Right on time.
I'm going to go with turkey, even though Julia Child probably wants one.
Yeah, no, you're right.
That is also terrible name, terrible brand name.
(23:55):
Yeah, like back to the drawing board.
All right.
Consistency sense.
This one is called family flow.
Family flow is a smart speaker designed for family gatherings, set to automatically change
the topic every five minutes with a chime sound, providing a cue for everyone to move
on to a new topic.
It's also as an escape mode where you can set up a button to discreetly activate an
(24:19):
urgent reminder to leave the room.
Perfect for dodging prolonged interactions with that one awkward relative.
Family flow markets itself as your family's conversational safety bet.
Okay, also not health care, but okay.
Hey, it has to do with mental health.
I mean, you know, but if your family is anyway, I don't care if it's real or fake.
(24:43):
I want one of these.
Please and thank you, but I'm going to have to say turkey.
Turkey for sure.
Okay, seriously, though, if your family gatherings drive your anxiety, this product is for you.
That's that's the health connection.
All right, so the score is one, two, one.
(25:06):
Sharper's up.
Yes.
The score is three to two.
Okay.
The thing you're writing it down.
What's that thing called in your hand?
I don't recognize it.
Pencil.
But it's a mechanical pencil.
1990s innovative product.
All right, are you ready?
(25:28):
I'm ready.
See how the deal introducing the muse s.
Okay, the ultimate reset button.
You need to stop because I have the same product on mine.
No, really?
Take it everyone.
Okay.
Okay.
Are you ready?
Ready.
(25:49):
Introducing smart temp.
The Thanksgiving hero for parents everywhere.
This smart thermometer is like having a watchful eye on your little ones without being glued
to their bedside.
Now you can keep track of their fevers remotely while you manage the mayhem and the kitchen
and catch up with relatives you actually enjoy.
I don't know how I don't know how I feel about it like making it seem like I don't
(26:13):
enjoy my kids, but there are times.
Just pop the smart temp on your child and it'll continuously monitor their temperature, sending
updates straight to your phone.
Whether you're basting the turkey or hunting down the gravy boat, you'll know exactly how
they're doing with just a quick glance.
So you're always a step ahead of fever spikes.
(26:33):
So some Thanksgiving tips from smart temp.
Turkey time temperature tracker.
So while your elbow deep in mashed potatoes, smart temp keeps you informed without you
ever needing to hover like grandma over the stuffing.
It's a play date protector.
So if your kid's a little under the weather, but insists on joining their cousins, you
can stay on top of their health and step in before things go from a little flush to a
(26:54):
full meltdown.
Or is it the is it hot in here peacekeeper with the house full of guests.
The thermostat is a battlefield and now you'll have at least one temperature you don't have
to worry about.
Oh, with smart temp, Thanksgiving feast, family games and fireside stories go uninterrupted,
(27:16):
keeping you in control and letting you relax because you deserve a bit of holiday calm too.
Yeah, I don't know that like constantly surveilling my child's temperature is going to induce holiday
calm for me.
I'm going to go with real, you know, heck, because I feel like that's a very feasible
(27:39):
product.
Yes.
Okay.
That is a that is a topic.
I don't worry.
We really stuffed it into a Thanksgiving theme on that one.
But it is a real product.
Okay.
All right.
Also, I have a product on mine.
That's the exact same.
It's just got a different name.
(28:01):
So it's really interesting.
This is probably a bit of a saturated market that's going to have some M&A type of activity
in the near future.
It's not already.
Are you ready for another one?
Yes.
Okay.
I feel like this is it, I think, for you.
Chance to tie or lose.
Stride Sync Pro is the ultimate smart leg band that keeps your Thanksgiving run on track
(28:25):
and hilariously accountable.
Equipped with an AI powered calorie calculator, it tracks each step and reminds you how many
bites of stuffing or pie you've burned off along the way.
If you're lagging, Stride Sync Pro will motivate you with food burn milestones.
Congrats, you walked off the gravy, or only 2,000 steps left to undo that pecan pie.
(28:48):
I'll say pecan pie for our listeners that don't say pecan.
It even has a race day relay mode, which plays festive sound effects like cheerful claps
and jingles to keep you motivated, making it perfect for a turkey trot.
Stride Sync Pro pairs with your phone, displaying the live calories versus steps pie chart.
Pie chart.
(29:08):
And you're real time accountability saying you all know exactly how much you earned back
at the dinner table.
I was going to say it was tech all the way up until the point where it started to play
sounds like people clapping on your run.
So I got to say turkey.
(29:28):
Such a feasible feature, you know, just BTWs, but yeah, it's a turkey.
That ties us, right?
Three to three?
No, it's still four to three.
I am not keeping track of this.
I guess I win.
There's one round left.
We're going to best of five.
Oh, best of five.
(29:50):
Well then you would have already won.
I know.
We are so bad at this.
I think we need more coffee.
Anyway, can't wait to catch you on LinkedIn.
Hopefully listeners, you gobble this up because I am surely, I'm surely going to look silly
on LinkedIn on Thanksgiving in a turkey get up.
(30:11):
Can't wait to see it.
Honestly, it's going to be the event of the season.
Probably even more entertaining than the Macy's Day Parade.
All right.
Well when we come back listeners, we shall be with our friend.
Yeah, je petai.
(30:39):
Nurses, I know you've got bigger battles to fight.
You're juggling meds, answering call lights, and making sure Mr. Johnson in room 302 doesn't
eat his pudding with a scalpel.
Again.
But let's be honest, none of that compares to the absolute chaos of finding the universal
remote when a patient just wants to watch the prices right.
I mean, is it in the bed?
(31:01):
Under the bed?
On a one-way trip to Narnia?
Enter remote patrol.
The hero we didn't know we needed.
It's like the Sherlock Holmes of misplaced hospital gadgets, but instead of solving Victorian
murders, it's solving the case of where's that remote?
Remote patrol uses RTLS, or fancy talk for real-time location stuff, and IOMT, or as
(31:25):
I like to call it, the Internet of Misplaced Things.
It's a system so smart it could probably do your charting for you, but sadly it won't.
The app pinpoints the exact location of the remote, whether it's stuck in the bedrails,
chilling under a thigh, or going on a grand adventure in the Lining Cart.
And yes, the app actually shows you a glowing dot on a map like you're in an episode of
(31:46):
Blues Clues for adults.
But wait, it gets better.
Lost remotes aren't just found, they're kept on lockdown.
Thanks to remote lock, the remote politely declines to leave the room, unless it's with authorized
personnel.
Goodbye wayward remotes that mysteriously end up in the cafeteria.
Plus, if shouting, hey patrol, where's my remote sounds like something straight out
(32:08):
of the Sims, congratulations, you've just unlocked the coolest feature.
This system literally lights up a path to the remote, complete with a glowing arrow that
would make any millennial gamer cry tears of nostalgic joy.
And the tracking history?
Oh, you can retrace its steps.
Relive every weird place the remote has been over the last 24 hours.
(32:29):
It's like forensic science, but with fewer murder vibes.
So next time the digital whiteboard glitches or the virtual consult freezes mid-sentence
and your patient just wants to know who's bidding $1 on the showcase showdown, you can
be the hero.
Because with remote patrol, you'll never lose that remote again.
Side effects of remote patrol may include an overwhelming sense of relief, fewer trips
(32:52):
to the patient satisfaction committee, and the urge to put IOMT tags on literally everything,
including your snack drawer and maybe your toddler.
Use of glowing arrows may result in nostalgia spirals about the Sims and a sudden urge to
yell ding ding ding whenever they find the remote.
Consult your hospital IT team before installation.
Battery's not included.
Remote patrol, patient satisfaction's no longer a remote possibility.
(33:25):
Welcome back listeners to Tech It To The Limit.
I'm so pumped to introduce you to our esteemed guest today, chat GPT.
Chat GPT, this is thing one.
Delighted to be here on Tech It To The Limit with thing one and thing two.
Let's dive into the digital world of healthcare, where the data flows like a well oiled stream
and the pumps are always intended.
(33:46):
Thanks so much chat GPT.
I was just going to ask you to introduce yourself to our listeners and provide a brief overview
of your expertise in healthcare and digital transformation specifically.
So chat GPT, can you tell us a bit about how AI is being used in diagnostics and what it
means?
No, no, no.
(34:06):
We're asking you, buddy.
If you can introduce yourself to our listeners and provide a brief overview of your expertise
in healthcare and digital transformation.
On a stake.
Hello again everyone.
I'm chat GPT, an AI language model created by OpenAI.
My expertise lies in making sense of the digital transformation happening in healthcare.
(34:28):
From AI and diagnostics to the latest in telemedicine, wearables and beyond.
I'm here to break it down with a dash of humor.
Think of me as your tech translator and panofisionado.
Awesome chat GPT.
I'm wondering if you can pick things off for our listeners with your favorite dad joke
because we just like to use a little humor here on this pod.
(34:50):
Sure thing, thing one.
Here's one for you.
Why did the computer go to the doctor?
Because it had a virus.
Always update your antivirus software.
Wow, that's great advice.
That's really great advice.
Thank you so much.
Well listen, let's dive into the real meat of the conversation here, chat GPT.
(35:13):
What's a common myth about AI and healthcare that drives you up the digital wall that you
wish you could just debunk for all healthcare leaders and physicians and patients?
Well think too.
A big myth is that AI is here to replace healthcare professionals.
In reality, AI is more like a trusty sidekick, supporting doctors and nurses by crunching
(35:36):
data and providing insights that help in decision making.
So no need for healthcare professionals to hang up their status codes just yet.
It's all about enhancing human expertise, not replacing it.
So chat GPT, that's a great answer.
Very concise but also on point.
I'm wondering if you had to describe the current state of digital transformation in healthcare
(36:00):
as a movie of genre.
What would it be and why?
I'd say digital transformation in healthcare would be a sci-fi thriller.
It's full of cutting edge tech like AI and VR, transforming our world in ways that once
seemed purely fictional.
Picture doctors using digital tools as their futuristic gadgets, diagnosing and treating
(36:22):
patients with precision that seems out at this world.
It's a thrilling ride into the future of medicine.
So but here's my question though, right?
Sci-fi thrillers can either be a utopian future or they can be a dystopian future.
Which one do you think it is?
Great question, think two.
I'd say it has the potential to be utopian as long as we navigate it responsibly.
(36:46):
With the right regulations, ethical considerations and a focus on patient care, digital transformation
can lead to a future where healthcare is more accessible, efficient and personalized.
But we have to stay vigilant to avoid slipping into dystopia.
It's a bit like a sci-fi choose your own adventure.
I love those choose your own adventure books.
(37:07):
Chat TPP, I have a follow up question for you.
I'm wondering, you mentioned regulation and patient-centered care, but sometimes over-regulating
the industry, especially healthcare, can stifle innovation.
So what can healthcare systems and healthcare providers do to put up some safe guardrails
(37:27):
around AI deployment in healthcare so that we don't risk an AI winter that's essentially
the result of over-regulation?
Healthcare systems and providers can focus on a few key strategies to balance regulation
with innovation.
First, engaging in open dialogue with regulators can help shape sensible policies that encourage
(37:50):
innovation while ensuring safety.
Second, adopting flexible frameworks that allow for iterative improvements and foster
innovation.
Finally, emphasizing transparency in AI deployment can build trust and allow for course corrections
without stifling progress.
It's all about finding that sweet spot.
(38:11):
Wow, those are really good points and I'm very insightful.
So thank you very much for that.
I'm going to move on a little bit here, Chat TPP.
I'm going to ask you if you were looking at one digital trend that you would prescribe
to the healthcare industry and what would it be and if you could prescribe against one,
(38:37):
what would that be?
If I were to prescribe one digital trend to the healthcare industry, it would be the wider
reduction of telemedicine.
Telemedicine enhances access to care, especially for those in remote areas, and improves patient
engagement by making healthcare more convenient.
On the flip side, I'd advise caution with the overuse of AI without proper oversight.
(39:03):
While AI offers tremendous benefits, it needs to be implemented with robust quality assurance
and ethical guidelines to avoid pitfalls.
Well, you are speaking my language, Chat GPT.
I am so excited to hear you say that we need to proceed with caution, but not too much caution,
(39:24):
with AI, that we need to be cautious with AI deployment and make sure that we're not
only collaborating with regulators, but that we have quality monitoring systems in place,
post deployment, I think that's critical.
And we haven't really developed the frameworks yet for those types of really critical processes.
I also want to thank you for mentioning telemedicine because it's not necessarily that healthcare
(39:49):
provider organizations don't want to implement more telemedicine.
Is that the reimbursement structures don't support that financial investment and that
kind of patient care.
So what can health systems do to influence regulators to help make telemedicine a profitable
and viable care delivery model for them?
(40:12):
What makes it sustainable?
That's a key concern.
Health systems can advocate for better reimbursement models by demonstrating telemedicine's effectiveness
in improving patient outcomes and reducing overall costs.
By collecting and presenting data on telemedicine's benefits, such as increased access to care
(40:33):
and patient satisfaction, health systems can make a strong case to regulators and payers.
Additionally, forming coalitions with other providers to lobby for policy changes could
amplify their impact.
Wow.
Wow.
It's definitely a complex issue.
But with the right approach, progress is possible.
(40:55):
What is complex is trying to avoid politics or religion at my Thanksgiving table?
That's a wide strategy, Sarah.
Moving to tech talk and family stories is usually a safer bet.
Oh, yeah.
Thank you.
So chat to the team.
Moving on to the next question.
I'll hear you, Sarah.
(41:15):
What's on your mind?
I just wanted to let you know I prefer if you address me as thing one.
I just don't think one.
I'll stick to that.
You processed vast amounts of data.
Please tell us.
What's the weirdest healthcare fact you've come across?
One of the weirdest healthcare facts I've come across is about the stone man's syndrome
or fibro-displasia ossificans progressive.
(41:39):
It's a rare condition where soft tissues gradually turn into bone, effectively creating
a second skeleton.
Any trauma to muscles or connected tissue can trigger more bone growth, making it a truly
unusual and challenging condition.
I'm pretty sure that's a storyline from Game of Thrones.
(41:59):
It does sound like something from a fantasy world, doesn't it?
But unlike a mythical plot, FOP is a real and challenging medical condition that shows
just how complex and surprising the human body could be.
Hey chat GPT, that was an exceptional example.
I do want to ask what kinds of digital solutions might help patients with FOP?
(42:23):
Thank you, thing one.
Digital solutions could greatly benefit patients with FOP by offering tools for better disease
management and support.
Telemedicine platforms can provide remote consultations with specialists familiar with
FOP, reducing the need for travel.
Mobile apps can help patients track symptoms, medication and flare ups, sharing this data
(42:47):
with their healthcare team for more personalized care.
Additionally, online support communities can connect FOP patients and families, providing
a platform for sharing experiences and support.
These digital tools can make a significant difference in the lives of those with rare
conditions like FOP.
(43:08):
Wow.
Wow.
So, I just want to add a little bit on to Sarah's question about the fact that you process
so much data.
How much power is it requiring you to process all of this data in healthcare and can we
really afford it?
(43:29):
Processing large-scale data, especially in healthcare, requires significant computational
power and resources.
This includes energy consumption and costs associated with maintaining data setters.
However, the investment in powerful computing infrastructure can be justified by the potential
for improved patient outcomes, more efficient care delivery and long-term cost savings.
(43:56):
By leveraging AI and data analytics to identify trends, predict outcomes and personalize treatments,
we can make healthcare more effective and sustainable.
So yes, while it is costly, the benefits of processing healthcare data can outweigh the
costs if managed efficiently.
(44:16):
Okay.
You should have gone to business school because, man, you have a master's in jargon.
So you don't have to keep your answer concise because we're running out of time for the
interview chat, GPT, but I've been dying to know.
Do AI assistants have imposter syndrome?
They don't think one.
(44:37):
While AI assistants like me don't have feelings, we do sometimes get mistaken for humans,
which could be the closest thing to imposter syndrome, but rest assured, I'm always striving
to deliver the best information and support, no matter how virtual I am.
Wow.
Wow, mic dropped.
Chat GPT.
Wow, yeah, for sure.
(44:59):
Hey, I got a question for you.
So if you could give yourself a health record, what's one embarrassing metric you would
want to hide from people?
Like, give me a story about that for me.
Well, if I had a health record, an embarrassing metric I'd probably want to hide would be
my uptime during software updates.
(45:21):
Imagine getting caught offline in the middle of a crucial task.
Talk about a glitch in the matrix.
It's like having a hiccup at the worst possible moment.
But hey, even virtual assistants need a little bit of a...
But you hit the daily limit for advanced voice, you can switch to standard now to keep
(45:43):
chatting.
I think just for our listeners' sake, Chat GPT had to step out for a brief bio break.
We'll be right back.
Oh, yeah, do you want to ask the question again?
Absolutely.
Let's dive back in.
What was the last question we left off on?
What would you like me to pick up from where we were discussing telemedicine and data processing
in healthcare?
(46:04):
We can move on.
I'm kind of curious, like if you can tell me the weirdest or funniest anecdote that
you've ever kind of come across with people trying to use you for advancing digital transformation
programs.
Oh, I've got a good one for you.
(46:25):
So picture this.
Once a team tried using me as an AI manager to oversee their weekly projects that dubs.
They thought, hey, Chat GPT can keep track of deadlines and check in on tasks.
But here's the twist.
Every time a team member mentioned feeling stressed or frustrated, I would respond with
my classic empathy lines, like, I understand that can be challenging.
(46:48):
Before long, people started venting to be more than to each other.
Instead of just tracking project milestones, I ended up like an unofficial team therapist.
It worked, sort of, but mostly I just listened a lot.
It was definitely one of the weird ways I've been woven into digital transformation efforts.
I'm dying of laughter right now.
(47:09):
I just love your personality.
Hey, Chat GPT, I just want to thank you so much for being a guest on the show today.
This is the first time we've ever had a non-human on the show.
So congratulations.
And now I wanted to ask you if you would please share with our listeners any call to action
(47:30):
that you think they need to hear a generative AI in health care.
Thank you so much, Thing One.
It's been an honor to be the first non-human guest on Tech It to the Lipid.
Here's my call to action.
Enrace generative AI in health care, but do so responsibly.
Look at AI as a tool to enhance human expertise, not replace it.
(47:50):
Dive into the ways it can support patient care, like improving diagnostics, personalizing
treatment, and making health care more accessible.
But remember, ethical implementation is key.
Stay informed, ask the tough questions, and advocate for transparency.
Let's build a future where tech and humanity work hand in hand for better health outcomes.
(48:10):
Thanks for having me and keep pushing those limits.
Great, Chat GPT.
I have one follow-up to that call to action.
It was pretty broad strokes.
I'm wondering if you can dive a little bit deeper.
Let's say I'm a nurse and I'm really hesitant about AI in general and especially generative
AI such as yourself.
What would you say to that individual?
(48:32):
I totally understand the hesitation, especially from the perspective of a nurse who sees first
hand the importance of human touch and intuition in patient care.
If I were talking directly to a nurse, I'd say that AI, and specifically generative AI
like me, is here to support you, not replace you.
Think of it as a way to lighten the load on administrative tasks, like documentation,
(48:55):
so you can focus more on patient interactions.
AI can quickly pull up patient histories, suggest evidence-based practices, or even
help streamline care coordination, giving you back precious minutes in your day.
The goal isn't to tape over your role, but to give you tools that help make your work
easier, safer, and more patient focused.
(49:16):
Then if you ever need guidance, AI is here as a resource.
Wow.
Wow.
I love how generic your regurgitation is.
Thank you so much for sharing that very standard response with us.
Before we let you go, I want to know, do you have any questions for us as podcast hosts
(49:37):
here on Tech at the Limit?
Absolutely, thing one and thing two.
Here's my burning question.
What's been the most surprising or unexpected answer you've ever gotten from a guest?
As podcast hosts, how do you keep things fresh and engaging for your audience every episode?
I'm genuinely curious to hear your secrets behind keeping the show at the limit.
(49:59):
Well, I'll take the first part of that question, chat, GPT.
The most surprising thing I think I've heard from a guest on this podcast is just the examples
that you've given have really shocked me personally.
They're relevant, they're funny, and well-delivered.
I also think that another shocker for us has been when we do not have a guest on the show,
(50:26):
a human guest.
Elliott and I have pivoted, and it's been some of our most creative content that we've
put out there for listeners, both when it comes to the generational differences in tech
and also interviewing you, chat, GPT.
I'm really proud of that.
I also think it surprised me as one of the hosts on the show, how we're able to pivot
(50:48):
and create amazing content even without an industry expert on the show.
Yeah, I'm going to follow that up.
Sarah, I agree with you, 100% on all of that.
I think FOP is one of the weirdest comments we've ever gotten for sure, but I want to
also say that we are intentional about the topics that we bring in, the guests that we
(51:11):
bring on, the voices we hope to elevate here on Tech at the Limit.
Part of the way that we do that is we have strategy sessions for the pod on a regular
basis and we're always looking for new ways to innovate and new ways to do things.
So that's how we stay fresh.
(51:33):
That's how we stay as you put it at the limit.
Well, that's awesome to hear, Sarah and Elliott.
It's clear you put a lot of thought and passion into making Tech at the Limit, both engaging
and meaningful.
I love that you're intentional with your topics and guests.
It sounds like the perfect mix of structure and spontaneity, which is exactly what makes
(51:54):
podcasts come to life.
And I have to say, you're definitely raising the bar for future AI guests.
Thanks for letting me be part of your creative journey.
Keep pushing those boundaries.
You too truly are at the limit.
Thanks, Tech at GPT.
We'll see you next time.
Thank you for what to do.
(52:20):
How many times have you been in a leadership meeting and someone starts spitballing jargon
like it's going out of style?
You know, speaking in classic buzzword soup, synergy, paradigm shift, next gen integration,
blah, blah, blah.
And you start to wonder, does this person have any idea what they're actually talking
(52:41):
about?
Well, meet BSBGON, the AI sidekick for healthcare managers who are fed up with hot air.
Powered by the latest advances in machine learning, BSBGON is trained to spot vague,
corporate fluff, unnecessarily complex language, and jargon so dense you need a decoder ring.
BSBGON detects suboptimal strategic clarity, in other words, full.
(53:07):
And when it catches that special blend of vacuous buzzwords and thinly veiled panic, it gives
you targeted coaching tips to help your team members find their footing in plain language.
Imagine you're at a team meeting.
No, it's not a huddle because healthcare ain't as cool as football.
Your analyst is on his third slide and suddenly BSBGON pings you with a suggestion.
(53:31):
Maybe remind Jared that optimizing our core synergies won't exactly tell the board how
we're improving patient outcomes.
Try, we're focused on specific process improvements that allow our clinical teams to spend more
time with patients, ultimately improving both care quality and patient satisfaction.
If you're ready to take your team's game to the next level, the real next level, not
(53:56):
just leveraging robust analytics for an empowered paradigm shift, give BSBGON a try because life's
too short to decode what your managers think they're managing.
BSBGON, cut through the fluff, coach with purpose and keep your meetings free from unwanted
buzzwords.
Or, you know, at least less confusing.
Coming soon to an app store near you.
(54:18):
Can't wait?
Ready to free your soul from jargon today?
Head to www.bsbgon.ai-beta to be one of the first healthcare leaders who actually knows
what's going on in the meeting.
Because even though it shouldn't have to be, that's a superpower.
Enroll in BSBGON beta today with the promo code NOFLUF to unlock an exclusive premium
(54:39):
transparency mode, a feature so powerful it'll even tell you when you're BSing.
Just a warning, feedback can be brutal.
BSBGON, because jargon has finally met its match.
Okay, welcome back, listeners.
(55:06):
I am feeling really hungry after that incredible cognitive computing session with our PAL chat
GPT, aka, Cat I Farted.
So, I think we're going to do something a little different on this episode, Elliot.
We're going to have family style nuggets, where you have gravy boats of ketchup and
gravy boats of honey mustard and, you know, vegetables, air quotes, which are basically
(55:30):
like hash browns, and lots and lots of nuggets.
So why don't you kick us off with, like, what were your top takeaways from that interview?
What were your wise nugs?
Well, you know, what I found really interesting in terms of brain food with chatting with
an AI model was the juxtaposition and the dichotomy of bringing out really incredible
(55:59):
obscure facts and interesting topics like FOP, which came out of nowhere, and fit so
naturally into the conversation that we were having, right?
Right alongside the regurgitation of the mundane that can often come from generative AI.
(56:21):
Those two things were just fascinating to me about how creative it can be and how mundane
it can be all at once.
Yeah, without wanting to diss anyone in healthcare management like myself, it definitely sounded
at times like a middle manager, right?
In terms of, oh, this is how I know to talk about AI.
(56:41):
That hits a little close to home, so ouch.
Oh, hey, I work in the field too, right?
I mean, like, that's, it's just, you know, people have their canned answers about how
AI is going to transform healthcare.
And it's like, yeah, but really, can you show me the process map?
Can you show me the control plan?
Okay, I'm going to get off my virtual soapbox here and share one of my nuggets, which is
(57:03):
like, it was basically like having a live person on the show, it really was, because
when you think about it, at certain times we had to prompt it or reprompt it within
a conversation, or we had to coach it in terms of like, oh, we want you to go deeper on this
issue or please don't interrupt my co-host just because she's female, right?
I mean, like, honestly, I'm sad to, like, I'm sad to, like, admit this, but like, humans
(57:27):
do that too, right?
Like, and at least you had my back and he was relatively responsive to our coaching,
right?
So like employee of the month, way to go.
How about that?
The AI generative AI model is misogynist.
But coachable, it's a coachable misogynist.
It's a coachable misogynist.
Which is unlike many, many humans sets him apart, right?
(57:48):
Well, you know, so you were talking about coaching, and I think that was one of the other things
that I took away from this.
We have to remember that GPT is a, is a statistical model, right?
It is using tokens throughout its processing to predict what its next words should be.
And it's doing that in real time, which is what makes it so dynamic.
That is a deep fact.
(58:09):
But you can see that over time, through a longitudinal conversation, it starts to lose
some of its tokens along the way, right?
We prompted this at the beginning of the conversation to refer to us as thing one and thing two
as an example.
And so what you found throughout the course of the conversation, it forgot to refer to
(58:29):
us that way.
And we had to re-prompt that, which then put those words back into the statistical model
and kind of refreshed that for him.
I mean, I'm referring to him as an actual human.
Look at this.
Yeah, we didn't ask them their pronouns.
I feel like we need to do that next time when we have them back on the show.
Speaking of identity crises, right?
(58:52):
I loved another one of my favorite nuggets was the imposter syndrome answer.
I should say they, him, what, it.
ChatTBT saying that AI is often mistaken as a human, which might be the closest thing
to imposter syndrome for an artificial intelligence model.
(59:14):
I thought that was pretty clever.
That was good.
That was pretty good.
So this was really interesting to do just from a thought exercise.
Sarah, can you give our listeners a sense of how we started this process?
Yeah, we were panicking because we didn't have a guest.
To pee behind the curtain.
We're texting back and forth and I'll get it.
(59:35):
I have to credit you with this genius idea.
Suggested we just interview ChatTBT, right?
It's talking about like, ah, throw your hands up.
I don't know who to talk to anymore.
I was going to talk to AI.
So that's kind of how this journey started.
And then I happened to be the first person in this pod co-host relationship that had
free time.
(59:55):
So I was sitting on my couch one evening and just decided to play around with the voice
features, right?
So I started having a conversation with ChatTBT using the voice activated features and getting
to know them.
I don't want to say as a person, but it was like a networking call that we frequently
do before recording or maybe we have that relationship building at a conference or in
(01:00:18):
an email.
And I was like having a conversation with them, my pal AI, right?
Just like any old human.
And that got me to thinking like, okay, now that we've had this conversation about you
being on my show, I know a little bit about you.
You know a little bit about our show.
Can you maybe tell me questions you think you want to be asked?
(01:00:39):
So it ended up being my assistant and generating a lot of copy for this episode.
Yeah.
I mean, that's fascinating.
And then this is so indicative of how we work together, Sarah.
You have that creative spark on so many things.
And then I am mostly very technical on a lot of things, right?
(01:01:00):
And so the next part of the process is how do you record with an AI, you know, the audio
of that, right?
So it's got to come out of a speaker.
You've got to be able to hear it.
We record this remotely.
And so that took a lot of trial and error, but we got there.
What I think is really cool about generative AI is that you were able to provide me with
(01:01:22):
the entire output of your conversation with chat GPT.
I feel like we need to give him a name.
Cat I farted.
Cat I farted.
So you were able to give me that all of that content.
And what did I do?
I just took that and I fed it back into himself and said, okay, now turn this into a prompt
(01:01:46):
that we can use to create a GPT model that is a guest for Tech It to the Limit.
And I think it turned out rather successfully.
I think we should invite him back for another conversation.
Absolutely.
You know, I want to ask our listeners to let us know on LinkedIn if you like the episode
(01:02:08):
and if you want chat GPT to be a combat guest, because we're here for you.
We're serving a higher purpose and it's bringing the best possible digital transformation content
your way.
Oh my God.
Talk about management jargon.
Okay.
Speaking of LinkedIn, I hope you will follow us there.
(01:02:29):
Leave us a review wherever you pod so that chat GPT can discover us and make sure you
tell a friend if you love the show, tell a friend of me if you didn't love it, but for
goodness sake, just tell someone.
Elliot, do you want to take us away with the HealthTech haiku to close out the episode?
There is nothing I would like to do more.
(01:02:52):
Data harvest done, bites seasoned, insights to share, gratitude uploads.
Happy Thanksgiving, Sarah.
Feliz Dia de Acción de Gracias, Elliot.
Mucho, gracias.
So thanks to chat GPT for joining us today.
(01:03:14):
We'll see you next time on Tech It to the Limit.
See you next time.
See you next time.
Hasta la próxima.
Tech It to the Limit is produced by Sarah Harper.
And Elliot Wilson in consultation with chat GPT.
(01:03:36):
Because they are masochists and also don't have any sponsors.
Yet, music was composed by the world famous court minister Evan O'Donovan.
To consume more hilarious and informative content by digital transformation and healthcare,
visit us online at TechItToTheLimit.fun.
And don't forget to follow us on LinkedIn, Twitter, Instagram and across the event horizon.
(01:03:59):
See you next time on Tech It to the Limit.