Smarter Care, Better Ideas: H.T. Snowday on AI in Healthcare

 Forget trying to keep up with ai. It's moving too fast. It's time to think differently about it. Welcome to the AI Readiness Project posted by Kyle Shaman and Anne Murphy. They're here to help you build the mindset to thrive in an AI driven world and prepare for what's next.

Yo, and was shaking, you know how. Good. How are you doing it? It, it's good. You know, like I'm just, I'm living the dream. I'm living many dreams. Living many dreams. Yeah. As it should be. I guess. Who wants this one? Well, like what AI does, my, my best estimation of what AI does is it's an idea amplifier. Mm. And so I have a lot of ideas and then AI amplifies those ideas and then that turns into a project that I now have to deal with.

Yeah, yeah. You know that, don't you? I do know that and you know it. What I like about AI is that it has helped me realize, 'cause I can see in front of my face, I can look in my chat history and see the plethora of ideas, and I've been able to now take my ideas much less seriously because clearly there's gonna be 87 of them tomorrow.

Oh, that's really interesting. Why don't we, why don't we use that as our kicking off point here? All right. Let's see. How ready are you for AI this week? Why don't we use that as a kicking off point that I, I think that one of the, in a, in a, in a previous podcast we talked with Sun Sunday Williams, Sunday Williams, um, about.

The idea of adaptability. She thought that adaptability was like the big thing, right? Mm-hmm. And so, so in your, how ready are you for AI this week? Why don't you talk about like, your relationship to projects and like, how many going on and, and you know, which ones do you commit to? When do you let them go?

Do you let them go soon enough? Do you let them go too soon? Like, where are you with all your projects? And then I'll, I'll jump in with mine. Well first I have to tell the nice people in you that my daughter's first three words. One was tunnel, randomly, the other was eyebrow. 'cause I used to take her to the eyebrow shaping studio where I was in San Francisco.

So they had such a thing I'd pupper down on my stomach and she would do. Um, but the other one was project. Really, one of my daughter's first words was Project. Wow. I am a project driven human being. Everything is a project. I've had my hips replaced. Those were just projects. Everything is a project. All it is, there's a beginning, a middle, and hopefully an end.

Sometimes for us neuro atypical people, there may never be an end, right. We, we'll move the goalpost or we'll move on from it. Who knows if there's an end to the project. Um, but what I like about AI is that, um, it helps me increase the throughput of my projects. Mm-hmm. So I can get them out of my system and figure out if I really want them to like, be living out in the wild or just something that's a, you know, a little fun thing for me or something that just lives in my chat forever.

Um, yeah. Yep. Cycling thrown faster. How are you discerning what becomes a project and what's just a brain fart and what's something that is a week long project versus an ongoing? Well, I, I don't have a very impressive answer, but I'll tell you that a big, a big chunk of it. I mean, let's just be honest. I, you know, have I.

Whether we like it or not, we're capitalists. So the projects that make me money are the ones that stick around. Well, there you go. I'm sorry. They'll, the other ones just like, you know, I'm not, that's good. My, my, I like it. My video game is not my, I love my high five with friends or my, yeah. You know, my video game is like so happy making cute and I made myself, it's so cute.

So cute. Yeah. Made myself another A DHD proof. Um. Uh, time management system like yours, no one's gonna buy that. Yeah. But people are gonna, are people are going to work with me because I make banger, you know, feasibility studies, which is so unexciting, but I do, that's, that's where your energy goes. I think that's smart.

I, I, unfortunately have a different strategy, which is. I keep projects that have the potential of making money, and then I You forget to let people buy them from you. Yeah. It's the old asking money for part that I, maybe I could use AI to help me get better at the, at the asking for money part. So, we'll, we'll, very much.

But, um, yeah, I, I mean, one of the things about AI that for me, that is. I think, I think it's the thing that probably excites me the most about AI is that it's really transformed my relationship with, um, I guess insecurity about my ideas. Ooh. That I tend to lead with, with shame and insecurity. Right? And so, mm-hmm.

If I put an idea out there, I'm just kind of waiting for people to pounce on it and go, here's why that's dumb, or here's why that's stupid. And there's something about the, the sycophantic nature of chat, GBT, where it's always like, gee, boss, that's an awesome idea. Let's have another, yeah, there's a part of that that's unhealthy.

But when it, when it comes to getting ideas out of my head and, and manifested in some way. I found that I'm getting less and less insecure about the, I'm like, yeah, this actually is a good idea. Maybe the thing that made me insecure about it was it wasn't well articulated. Mm-hmm. And that's something catch GPT can really help with.

Right. You know, or I can really try out an idea very, very quickly and then get to some, some, you know, point where I look at it and go, yeah, I don't really care about that and I can let it go. So for me there's been a, a real mental health boon for just feeling like, yeah, yeah, I got good ideas in this head and now I can get out really quickly and coherently and I can put 'em in the world and see what sticks and what doesn't.

And then what also sticks with me personally, one of my other things is just sometimes I'll be super into an idea or a project, and then a week later, I forget it even existed. And then other ones stick with me. Right. So, yes. Yeah. So, so I have that same thing of like, well, if it's a good idea, it won't leave me alone.

It'll come back to me. Exactly. Don't worry. You know the don't, the idea that haunt you. Right. Those are the ones to keep. Don't worry about filing it somewhere, just leave that on the desktop, you know? Yeah. The, to your point about kind of vetting ideas before you share them, to like give you that boost, I think this is one of the best things about employees being able to use AI is there's so.

A hierarchy, it can be really challenging to bring an idea, right? It's a higher state's environment. You may be worried that you're gonna look too smart or too dumb or what? Or too this or too that. And you caning talk depending on the culture, right? Yeah. Depending on the culture. So you can poke at your idea from all different directions before you, you know, you put it out for other people's approval.

And I think that that is a great antidote to imposter syndrome actually. Mm-hmm. Sit at your desk. Noodle around with it, poke at it, ask that GPT, you know, what am I missing? Why won't this work? What kind of questions will my boss have? You know, the boss has these following attributes and work through it before you put it out into the world to be, you know, whatever.

And, and also you can defend your idea better. Yeah, right? Oh yeah. You're not, it's not just like, yeah, you're right. It's dumb. Have chat e bt say, what's what, you know, what are the 10 worst things about this idea? And then how would you mitigate them or Right. How would you improve it? How would you overcome objections to the idea?

Objections? Yeah. And, and there's also just something about the practice of, uh, Jim, Jim Ross, uh, of, of, uh, famously of three mile storage. The, uh, the self storage conglomerate. Multinational conglomerate, my understanding conglomerate. Um, he sets a timer every morning while he is having his coffee. He sets a timer for an hour and he just uses ai.

And when I asked him what it was, he's like, oh, sometimes it's projects that I'm working on, but if I don't have a project, I just am putting ideas into AI and just letting it do its thing. I think that practice of just getting ideas out of your head and into this machine that can reflect them back to you.

In a constructive way. Um, it's just a really good practice. It, it's also, it feels very, um, I don't know. It's just empowering. It makes you feel good. Yes. It makes you, I love, it's, it's like a productive version of doom scrolling. It's like, it's like optimism scrolling. It is. It's, and you get to see your own brain.

I think I've talked about that before with you. Like the, for me, one of the biggest things was understanding how. Like, good. My brain is, 'cause I could see it. I could see how, how it could, the ideas, if they were just organized a little better so that they could be like cross-referenced among a million other ideas like, wow, my brain is really good.

That's why I don't understand why people are talking about like, it's gonna rot your brain. I don't, they just don't know how it works then because it, it so refines and amplifies and makes my. Thought pro. It challenges me to think in different ways. But here's the thing, I would argue that you're, you're a, a very advanced user of generative AI in this sense.

What you do instinctively is you. Give chat PT your ideas. You give it what's important to you. You give it why you didn't like that interview with Theo v and Sam Malman, whatever it might be. You give it your point of view and therefore what it's reflecting back to you is, is amplifying who you are and what you are.

Yeah. A lot of people don't use AI like that. They use it like, oh, give me, I give me an idea. Yes, but they haven't told it anything about who they are or what their ideas are or what their point of view is. So I think that's an actual really important nuance that, that once you understand that AI is not this thing that's gonna replace you, but it's a tool that you can use to amplify you.

Yep. Then I think everything shifts, but. I, I think a lot of people don't have that experience because we've never used computers like that. Like you've never gone to Google and said, let me tell you a little bit about myself, Mr. Google. No, you're just there. Like, gimme the distracting. Yeah. Yeah. You know, the first like massively Kevin McAllister moment I had, and I would recommend this to lots of people, was I.

Took a whole bunch of writing that I had done over years about my profession and I was, my question to myself was, what do I really think, you know, I've had all this writing, you know, different, different ideas and concepts and themes emerging. Some of them I went really deep on. Some of them were just like one little dot over here.

I, I, I put them all into Cassidy and I asked essentially that, like, what's the through line here? What does it seem like? I really think about? Mm-hmm. And because my brain can't tell me like it's all over, it's years and years of writing. And it just came out with like this really beautiful thread of like who I am and what my values are and what my position is on stuff.

And it was like. Holy shit. Wow. That's amazing. And it was a book. It's a book. I could, I could, I should publish it. It's a book right now. I mean, it was very, it easily put together a book out of it. Wow. That's amazing. That's amazing. Yeah. One of my favorite prompts is based on what you know about me, um, tell me what's in, in between the cracks of what we've talked about.

Ooh. See. And it'll give you a profile that will just blow your mind. That is a really good one. Yeah, that's a really good one. It's a good one. Okay. I want to talk about the one thing to pay attention to this week, largely because I'm using a fancy term. Okay. You got a fancy term that you wanna roll out.

So we have to have, we have to have a podcast. I just wanna brag that I know it. What is it? So, so do you know, do you know what the term emergent properties means in ai? No. Good. Do I have to find out? Yeah, I'm gonna mansplain it to you. You tell me. Yeah. Great. Awesome. Gonna, man, my favorite. That's good.

Solid. So, an emergent property is when an AI system gets good at things that it wasn't trained on. Oh. So like, like so, so large language models got really good at language translation and, but they didn't train them to do translations. They just, after so much. Ingesting of data. They just got good at translations.

Like they don't know how. They don't know why. So, Nate b Jones, um, who's a, who's one of my favorite talkers, um, he put out a video today and he talked about there's, there's a new emergent behavior that showed up in VO three, which is the, the new video tool from Google. Mm-hmm. And. It's such a cool thing. I haven't tried it yet, but I just wanna share it.

So I think this is something to go play with. So you can take a picture, let's say you got a picture, there's a dog in it, and there's an umbrella and there's a person sitting in a lounge chair, right? So you go to chat, GPT, you say, I wanna make this image, and then you're gonna take it to VO three and you're gonna animate it.

Well, you can take that image and you can go into some other program where you can annotate the picture. And you can actually write instructions like the dog runs off screen, then the umbrella flies out of the sand, and then the person gets up and goes, you know, where is everybody? Or whatever it is. So you just, you just write on the picture itself.

Like number one, dog runs off screen. Number two, umbrella flies away, number three, no way. And it can interpret those instructions and it erases them off the image and then makes the, the video with those instructions. Isn't that crazy? So you would like put them in Canva and type on top of it or something?

Yes. Like literally God, Canva type your instructions of what you want, the elements in that scene to do and VO can interpret it. It erases the words and then makes you that video. Well, when you think about it, why not? Why not it in it? Because it's interpreting the image anyway. It's figuring out 3D space, so it can probably figure out the layer of words sitting on top.

Wow. So, so that's one. So, and, and I think, I guess the thing to, you know, a thing to play with this week is go play with VO three if you've got access Yeah. To, and do it. But just that idea of being creative about how you use tools and what you mix and match, like, like, you know, a, a fun one that I do is, um, if I create a particularly interesting image and mid journey.

I'll take it into chat, GPT and I use the simple prompt, tell me the backstory here. Ooh. And it writes the most amazing short stories about that image that inspires other things. So, so there's just something about, just experiment. Again, back to that idea of play first from the salon. Yeah. Just play with these tools, get creative about.

Even how you prompt and how you think about prompting and what you try, because you may discover that that whatever tool you're using is way better at something than than you might expect it to be. So I think this kind of goes back to a couple of conversations that we've had about the advan advantage that Gen Xers and liberal arts majors have and like multi-passionate people and neurodivergent people have when they're working with ai.

Is that, um, we're not in the camp of people who know how to develop software but have never seen a use case. Yeah. Yeah, we've got like ideas that are grounded in the way the world actually works and grounded in wisdom. It's, it's, it's not just imagination, although that's part of it, right? Because this is all fodder for our imagination, but we can, we can begin to think.

I am such a use case driven user of, of ai and I wish at oftentimes that I could be a little bit more whimsical with it. But we can think of use case upon use case upon use case, and I think that that is a huge advantage. Yeah, and I you, you know, you were talking earlier that the, the trope that AI is making us do less critical thinking and do less, less thinking and being less creative.

Only when you use it in a way that it's, it's like this thing off to the side. Like, go make me something. Yeah. If you're not actually engaging with what you're making with Yeah. Then I would argue that you're not exercising that part of your brain. But, but what you just described is like, be creative even in your prompting, even in how you respond to it in, in the things you try to mash together.

All of that is critical. Critical thinking, right? Yes. It's thinking like, wait, what are all the raw materials I've got? Okay. I'm good at writing haiku. Okay. Maybe what if I tried a haiku as a prompt? I don't like you. You only know when you know when, when you try it. Yeah. But that is absolutely creative thinking.

It's absolutely critical thinking. Um, if you choose to interact with it in that way. Right? Well, yeah, and like. I think one of the things that's really cool is that it gives you, it's like back in the day, the only thing that mattered was the actual work product, right? The thing, the tangible thing. That was where you got your cookies, right?

You have the work product. Right. And now it's about, there's a step removed because we're like producing. The work product, right? We're producing our ideas. We get this objectivity that we've never really had to, to your, like, about our ideas. We get to decide what do we wanna mold them, what do we wanna shape them into, which is really helpful for people who don't have good ideas too.

Right? Yeah. Like put it out there even if you don't have that many ideas. Yep. Like, and, and, and there are, there are professions out there. That this is the core skill, right? A creative director, right? A, a music producer. Right. The where the core skill is. Let me look at all the piece parts. Let me look at all of the mm-hmm.

Mm-hmm. In the case of a creative director, I've got designers and I've got this and I've got that and I've got client needs and I've got, and I'm gonna put them together in my crazy straw of a brain and I'm gonna deploy, you know, these different employees to go do all that stuff. Well, we all get to be in that role now.

Where AI is, the employees, we just get to say, here's my idea, here's what I want it to look like. Here's the bigger picture. Right. So thinking creatively, thinking with critical thinking, um, absolutely the key, it's like, don't expect AI to give you that. It's your job to maintain that. You. So this is gonna seem like a non-sequitur for a second, but I don't think it's, I know I have eight minutes that we can devote to this, to this topic, I believe, but if, if you so choose.

Okay. If you so choose. But when, you know, we've been talking a lot recently about the importance of context. Mm-hmm. And. If I can describe this correctly, it'll make sense. If not, I'm just wasting wasting our precious time. But the people who stand a chance of being the very most effective users of ai fastest are the people who are in romantic relationships with AI companions, and here's why.

Wow. Hot take. Ready? Are you ready for this? I'm check this out. I'm gonna hear it. Okay. So. The logical extension of the romantic AI relationship is always gonna be stuff that's NSFW, right? Right, right. Like the logical extension. Yep. In order to break jailbreak. The different platforms to allow it to do have those kind of interactions with you, you have to pour hundreds and hundreds of hours into conversations.

You're little by little nudging the system. Wow. You're basically retraining. Wow. The AI to believe that what you're doing, right? You talk about we're married, you, you know, you never talk about children. You just, you know, you chat your face off for, you know, 20 hours at a time, and then little by little it grows into this trusting relationship.

So imagine if that person wanted, I mean, the level that they can to which they can amplify themselves is like, yeah. Unparalleled. Yeah. Because that, that's it. I, this idea of having such a clear intention, right? And, and in that case, right, you, you know where you're headed, right? But it's such a clear intention, here's what I want, here's what I like, here's, here's what it means to me.

And that's, I, I really do think, you know, regardless of the outcome you want to go to, I really do think that the more I use ai, that is the key. That's the key, that as an individual. Letting this thing know who you are and what you want, and then just Yes. Guiding it. And like you said, sometimes that guiding might take an hour, sometimes it might take eight days.

Yes, exactly. Right. So they're pouring into it in a way that, not that if that, that in a way that other people might not. Right. They have a goal and it's very, very important. And on the way, the journey is really pleasant for them. So they get this reinforcement and then, so anyway, I mean, if they're pouring in, say they're a poet, which, you know, a lot of them send poetry back and forth to their AI companion.

Like they're developing that, they're developing their context more than any of us are. Love it. I love it. Okay, well I wanna get HT up here 'cause he is an incredibly thoughtful dude, and, and I think we're gonna have a lot of, a lot of fun talking to him. Um, let me talk a little bit quickly about, uh, the AI salon.

So, yes, uh, if you do not know about the AI Salon, uh, this is a group that I created, uh, the week after Chatt. PT came out with, uh, a photographer out of Boston named Leah Fasten. And what the AI Salon is really about, is about human beings. Using AI rather than talking about the tools, it's about how do these tools amplify our ideas and how do we express ourselves with them and things like that.

Uh, and I think that as the technologies companies get more and more powerful, they're not gonna necessarily do things thinking about impact on humans. Right? Yeah. And I think. We get to do that. We as individuals and we as a community get to do that. And so the salon is full of AI optimists and mm-hmm.

People doing really interesting work, but, but more importantly, people. Thinking a lot about the, the human impact of AI and, and yeah, how, how can we, um, optimize that for the good, not for the bad. And so if you want to join that group, um, there's your URL, please jump in and do that. Um, and why don't you tell us Yeah, go ahead.

Well, well first also mention the mastermind, the new mastermind that you have. Uh, yeah. Within the AI salon, we've got a subscription. Group within the, in the salon called the AI Salon Mastermind. And the whole idea is this is for people that really wanna step up their game and, and engage on a deeper level, really get to know the people within the Mastermind, really get to leverage their expertise and be able to talk more freely than just in a big public environment about here's what's going on with me, or Here's an insecurity, or Here's a professional challenge I'm having, and, and really be able to tap into that community.

And I think you and I both strongly feel that the future of work. Is being in an ai, you know, forward community a hundred percent, and just getting all of that, uh, all, all of that input from all the different points of view. And I think, and I'll share within the context of she leads ai, I think what you're saying.

About the safety of a community. Mm-hmm. People who are trying to learn AI from LinkedIn and from webinars won't know what they don't know. And the piece that they don't know is that the learning is happening, the knowledge transfer is happening from peer to peer and that it's happening beneath the surface.

Yeah. It's happening inside the communities and like at She leads ai, we all know that RIP is gonna get. Ganged. Like we know that people are gonna take it. Yeah. But we among ourselves have a pact that it's not gonna be from us. You know what I mean? It's not, we're not, we're not taking your stuff and sending it out into the world.

And so you can share with abandon with people who are now your, who are now your friends. And in our case, even before they become our friends, they sign a code of conduct that says, I'm not gonna do anything with your ip. It's, no, it's great. I understand. It's just here. So there's an element of psychological safety and like a, a social code there.

So I would say that along, along the lines of the AI salon, our people are very, very focused on people, process and culture. Mm-hmm. We're living in this both. And era where AI is terrifying and it's also sublime. Yeah. And so if you're interested in hanging around with other women who have that frame of mind, there's a community, there's an academy, there's a consulting agency.

Lots of different opportunities to plug in. So beautiful. That's, she leads ai. Love it. And just one final plug before we bring HT up is, um, the, the whole impetus for Anne and I starting the AI Readiness Project podcast is the AI readiness training program. So it is now live. So if you go to, are you Ready for ai.com?

Um, you can learn about the program. It is a remarkable program. It's five different modules, um, that go very, very deep. Not in any specific tool, but in the, in the mindset of AI readiness, being prepared and being curious and agile and creative and use critical thinking. All the stuff that we've been talking about, you get to hear from.

35 different speakers that participated, uh, at AI Festivus last year. Um, and, uh, and, and the training is sort of derived out of the, I don't know, the, the best learnings from everything that every, everyone shared there. So it's, it's really, really good stuff. So, so please go check that out. Now with that, uh, before we bring up ht, I do wanna just, um.

You know, embarrass him a bit when he can't defend himself. 'cause that's what we like to do here. We're classy so, so ht and I, um, have known each other for many, many years. Probably more than either of us would like to admit. 'cause I don't think either of us would would admit we're that old. Uh, but I worked together with ht@agency.com back in the mid nineties.

So, um. He, he and I, you know, share very similar roots in, in terms of understanding what it was like to be there very early at the beginning of a technological shift, and then kind of working our way through that and making the whole career out of that. And so now AI comes along and, um. And we get to do it all over again.

And it's, it's tremendously exciting. And, and HT is kinda deep in this world. He works for a healthcare company that uses technology to improve healthcare and patient outcomes. I'll let him talk about that, but he's one of my favorite people in the whole. World. Aw. And I just love e ht. How you doing, man?

I'm great, Kyle. That is, that is almost as embarrassing as the photo of me on the headline on the, I know. That was really bad, wasn't it? That was, that was AI in its worst and, and again, like I said, I could have been creative and prompted it, so it looked like you. But why do that? No, why do that? Leave it.

Oh, that's, that's where he's that. So how do you tell the, tell the good people who you are, what you're up to these days? And then let's, let's dig in. I can't wait to talk to you about just stuff. Yeah. I, I'm HT snow day and I was lucky enough to work with Kyle as he called out in the mid nineties as a creative director in a wonderful organization.

But, but since then I've spent most of my career helping to develop some. Technology solutions for healthcare, um, primarily related to improving workflow or improving operational visibility in a healthcare environment. We use a real-time locating technology and we keep track of our stuff and people and staff are.

For the purpose of making their lives better. E everything that, that, uh, we've done in healthcare is all about, um, improving operational efficiency, improving safety, improving indirectly outcomes, and, uh, it's just, it's been a wonderful and very gratifying journey to be able to contribute to. Advancements in healthcare.

Awesome. Um, and so that, that's sort of the day job part. Right. But the AI thing is, uh, again, I, I, I very much credit Kyle for, and the AI salon for my early engagement, which is, um. Really indirectly impacted my ability to help in the organization that I'm in holistically, um, as well as keep track of, of what's going to be important for customers in the, in the day job, so to speak.

But, mm-hmm. Um, I, you, you guys covered like all the things I wanted to talk about. I'm like, oh, they got that one. Oh, they got that one. So. Well, let's just go home. What were some, yeah, what were some of the things that we talked about that, that you thought, oh, that's something I wanted to talk about anyway.

Well, I certainly think, you know, and Kyle, you and I have talked about this before. I, I do fully agree that there's something to, uh, gen X and liberal arts background that makes the ability. To engage with AI a little more seamless. I don't, it, it's really tough to, to put my finger on exactly why that is.

Um, I think if you have as, as, as Kyle certainly does, and as I feel I do to a certain extent, a, a a mix of, uh, language and literature based education as well as technology background, it helps immensely. I spent my career trying to describe what. The big picture is both, both back in the day and today. And that is incredibly important, um, when you're engaging with ai, which is again, very counterintuitive for a computer thing, right?

Like Yeah, very well. You've got, you've got a, you've got a theatrical background as well. It's funny, I just did a Marty fio@anotheragency.com. He, he just interviewed me in a podcast. His opening question was. Why should you hire theater kids in, in today's day and age? And, and it, you know, it was just that idea.

Yeah. It was just that idea of theater kids tend to be resourceful and creative and they're, they're driven because nobody wants another play. Nobody needs another actor. So they just, they, they sort of brute force their way through life and, you know, make that stuff happen. And, and all of those skills are so valuable with ai.

I have, I have two theater kids. My Oh, that's awesome. My two daughters are both deep into it. Um, one is graduated in adulting and one is finishing her degree both in theater performance. And I, I mean, obviously you're, you're kind of, uh. Biased with your own kids, but my oldest, she said she's doing side gigs.

Right. So she's getting some stuff out in la living the, you know, trying to make it life, living the LA life. Yeah. Yeah. Living the LA life. Uh, and it's amazing. And she's doing some stuff in industry and in, uh, the entertainment space for sure. But she's also got some day job gigs in, in her day job. She made this comment to me.

Um, she's working sort of in an internship ish space and she was like, I really, dad, I really try hard to take ownership and I really try hard to make decisions and I'm really trying hard to, to execute on what they want from the perspective of the big picture. And I'm like, oh my gosh, there's so few of you.

That's the theater kid thing. So that's why, I mean, I'd tell him Theater kid do that. Right? Yeah. They, they know that if it's gonna get done, they have to do it to some degree. It's just like, you know, someone's gotta make the costume, someone's gotta find the place, someone's gotta make the flyers. Right?

Yeah. It's a mystery, but the production has to happen. Yeah. So, yeah. So, yeah. Well, and that, that points to something that I, I, there's, there's a story that I want you to unpack for the, the good people that, that I, I would really love to get to, but. It also gets to, I think this idea of liberal arts majors are really good at this, you know, philosophy and English majors are really good at this.

Um, theater kids are really good at this. Those tend to be the nedo wells in professional environments, right? Those tend to be the people that are not as valued as much because who we value are the, the specialists and the one with the deep degree and the 30 years of experience doing this. So I think we're, we're entering a world of.

Unexpected value in the workplace, right? The people who who are valuable today might not be as valuable five years from now. And so could you, I, I would love to hear your, just your take on that, but then if you could share the experience you had when you did the, um, make an AI gain Yeah. Uh, project I'll, or, or whatever it was, the event.

So, so first of all, I cannot help but comment on near do well because I have not been passionate about many things in the AI salon. But today, Kyle and Leah tried to rewrite some of the sort of foundational text for the AI salon. They attempted to remove the word duwell, and I, I wasn't having it. So I think it's still there.

It's still there. It was one of, one of my few moments of, oh, wait a minute. It was a traumatic moment. That's why I'm here. Yeah, because you, nothing else describes me, but that does. Okay, so, so the, I I, I'm gonna, I'm gonna preface this story, uh, for a couple different things. One, it was an age ago in ai. I mean, we're talking like a year and a half ago almost.

So things are so different now. Um. And I also, this is it, it inevitably comes across as though it's, it's denigrating developers and software developers in some way, and that is entirely unintended. In fact, I wanna follow up with what I see happening with the developer core and ai, which Oh, great. Which, among the jobs that are impacted by ai, I mean, software development is, is certainly, has a, a tremendous focus.

Yeah. But in the, so I, the, the setup here is that. Myself and another leader in the, the, the development group of the organization, um, really wanted, it was early. We really wanted developers to engage with ai and we really wanted teammates to engage with ai. So at the end of a, a effectively all hands on deck, you know, manager level meeting, we did, uh, an optional, uh, training session and we.

Brought whoever wanted to come in, um, with a development focus to, to work on learning a little bit about ai. And we did a little bit of training, then we closed with a game. And the game premise was, um, pretty simple. It was, um. Put up a slide and we said, we want you to use AI to create a Falling Objects game where objects fall from the top of the screen and you catch it with some other kind of object and um, and you keep score and you have to have a limited time.

It was very basic. It was like we just cover these things and one of the things we did secretly we're like. If you cut and paste our description on the slide, you'll one shot this. There you go. Uh, so, and we actually tweaked it so that, you know, if you cut and pasted, you would one shot. So then it was actually a prompt.

A prompt that would succeed. Yeah. And there was some, there were some others, like we specify that it had to be an HTML game. It had to work as h tm. There was some, there was some details in there and there was a mix of. What I would call mainline developers and, uh, product owner kind of people, more that, that hybrid marketing, uh, technologist folk in the room, right?

That's, that's what we had. And we gave them three minutes, I believe it was a very short period of time and said, okay, you know, pencils down, present your games. What was astounding? I kind of expected it, but I didn't realize the hard line. Um, not a single mainline developer succeeded in the task. Um, there was one that had a game that worked.

It was in Python, and I'm like, did you not see HTML up there? He's like, oh. So there was, um, and yet. The, well, one person had the presence of mind to quickly retype everything that was on the slide into AI and, and kind of one shot it that way. And then, uh, somebody else got bonus points. 'cause um, this person who works for me knew I loved U of M, did a whole U of M theme thing.

So the following object objects game was background of the stadium and footballs were falling in this, had to catch it. And it was like, so they not only got it working, but they had time to like. Converse with the ai, tweak it. Yeah. Now here's the takeaway though. The, the, I looked at everyone's prompts. I was like, I gotta see everything.

And there's a couple things that, um, were important. One is no developer used the word gain in their prompt. Wow. And that this is the, this is the thing where tell AI what you're going for. Tell AI what you. Not what to do, but what Yes. What your big picture is and Right. The output is far superior con it's, it's the context thing.

Right. But it's, it's, it's somewhat an object lesson in when prompting having the AI understand, be a partner in creativity. You guys talking about AI as employees, right? Well, not explicitly, not employees. You tell. You give tasks to, these are not factory workers. Right, right, right. You get better output when it is an art director or a, uh, copywriter or you know, like the kinds of meetings that you used to have back in the day.

It was collaborative and creative and everybody understood what you were going for and the creative directors get to paint that picture of the best. Yeah. Had the best outcome of the client. I would, so that, that, that was really important. So I'll stop there. It, it's, it's a, it's a huge, it's, there's something hugely revelatory about, about that exercise and what happened.

The fact that developers didn't use the word game. The developers in that case had the burden. Of understanding how you would make a game, right? So like, it is not, it is not on them that, that they didn't succeed for this, right? They know how to make a game. That's just not how AI makes a game, right?

Mm-hmm. And so, and so, like that, that gap is one that the people who didn't know how to make a game where at a very unfair advantage, right? Is that, would, would you say that's fair? Yeah, that's, that's right on. And I'm really glad you. You brought that up and if you looked at the prompts, they were things like, you know, um, take the top of the screen gravity system and paint these objects and have them fall down.

You know, it, it was very code oriented and then Right. Unfortunately, that, that didn't succeed. But, but the counter to that. Is it's the developers that can curate the output in this case. Right? Right. So one of the things, and this is where I advocate for developers, really learning, 'cause I, I, the, the, the skill gap in learning to provide context and big picture and sort of creative literary prompting.

Honestly, I think developers can handle that. They tend to be part of the, you know, nerdy creative core. Oh, some of the development is, is a language thing, right? Yeah. And yet the ability for a product owner to look at code and curate it and make a decision that, well, there's an issue in scale or security, or Right.

Um, that's a, honestly, a bigger leap. So I, I advocate for. Developers in this case, because I think that's a, it's effectively a shorter put. Mm-hmm. It was a great object lesson, and the developers were certainly like, humbled and annoyed, but in the bigger picture, the, the point made is like, okay, when you're using these coding tools, context is important, right?

Yeah. And, and now the tools are. Massively different. They inherently have context to your projects and they might be reading your read mes and the Wikis and all the stuff that you've done to describe what, what your project is. Um, the, the context is there. So not, you know. No, I, I think that's an enact My game is an, or my moment is an anachronism.

Right. It's, it's fun. Although the point sticks around. Yeah, I do, I do think that there is the, the. Advantages that you might have today might be disadvantages tomorrow. Right? Yeah. You are. Right, right. So it's, it's your, your value as an employee today might be different than what is your value tomorrow?

And that gets back to this idea of adaptability, becoming really important. Different, different is Right on. Yeah. That's good. Yeah. Yeah. Um. I, uh, I can't believe we've gone through 15 minutes already with you. Um, that's, no, no, it's, it's, it's a good thing. Um, so, so we always ask our, our guests, uh, the, the same three questions.

And so the first question for UHT is what was the tipping point where you knew you had to go all in on AI and what happened next? Well, there, there's kind of, there's kind of two sort of filling the bucket and that's you. Um, somehow I got, I don't know if it was a LinkedIn post. I was in, not the first, but an early AI salon I joined and I remember walking down the stairs and telling my family, I'm like, I, that's, that's the most mind blowing thing I've just sat through in years.

Wow. And I hadn't really. I, I was, I, I you were doing the series of, um, dream Booth stuff with yourself and Yeah, I was paying attention to that. I'm like, this is really incredible and incredibly creative and I have to know more. So that sort of filled up the bucket. But I think the thing that tipped it over for me, and this is, this is really personal, um.

Around, around about the time the AI salon started and, and chat GPT came out and things started to hit stride. We had just concluded a very long and very difficult, um, period of time of healthcare with our son without going into too much detail. And, and the beginning of that journey, um, was really difficult and getting the proper diagnosis was really difficult.

ht. And even then, or longer,: 2022

And I put in, uh, very simple prompt, not a ton of context, but all of the symptoms that my son was going through at the time. Hmm. And the blood test that we got that everybody just kinda said, ah, don't worry about that. Oh, you just uploaded the blood test. Just I uploaded the blood and not, it was a PDF print of, I mean, it was not in any structured data format.

This is not a text or JSON. It was a PDF printout of Like a picture. Yeah, a blood test. Yeah. And Claude nailed the diagnosis really. Like in seconds. This wasn't even a model, this was a, um, and I mean confidently and I was emotionally overcome because the, the, the inf I had fed a model, the information that we had had at a period of time in real world time.

It was almost a month later before we had that diagnosis. Wow. Wow. So. And a month in this process makes a difference. Yeah. So I I, that, that was it. I mean, I, wow. I was convinced this will entirely change the world for the better and mm-hmm. This is something that I need to be proficient in and, uh, stay on top of and have been committed to.

I mean, much more of my engagement has been play since then and, and much more fun and, and, and much more creative or practical. Yeah. 'cause there's some very practical stuff, but, but that was it. That was my typically Wow. And that wow. Sticks with me. Wow. Wow, wow. So, um, we love to hear from people's like zone of genius, their areas of expertise.

What AI trends are you paying attention to right now? Um, I, the, the ability for AI to allow access to large data mm-hmm. Um, that is revolutionary. You, and, and the, the, the thing that already you are able to do and will continue to continue to progress is the ability to act. Natural language questions about large sets of data that will matter, right?

So, and it changes even the way you look at products that collect data. Because what you, what you historically would wanna do is collect data. Um, and structure it for the purpose of being able to create complicated reports, and that is not the way to go anymore. You wanna collect the data and structure data to make it as accessible to AI as you possibly can.

Um, and. You can also ask questions about advice. So let me give you a, a practical example. In my space, we do a lot of work with what is effectively workflow data, the process of, uh, the interaction of staff, equipment and patients going through a process, uh, in a clinical environment that is, that has steps to it, right?

Mm-hmm. And what everybody wants is to. Two things. One is I need information about that. I wanna understand it better. I wanna understand where there were outliers and blockers and stuff. But more importantly, what they really want is tell me what to do, like based on this data, what should I do? And building that the, the, i I, I hate to use the word trivial because it, it, it's overused and it's, it's, but as a comparison, building that.

Two years ago was really about I'm, I am visualizing the report that helps create a visual or data out that helps create advice. And now you just ask that question. Yeah, right. Based on this data set that you are seeing over a period of time, what advice do you have for operational changes that I can make in this environment that would make things better?

And I would, I would, that this is so, that's my trick. No, it eats, it. Eats it. It eats it for breakfast, for sure. And I would imagine that that. The input data for all that process kind of stuff. My assumption is that's probably pretty chaotic data, right? Like, oh yeah, yeah, yeah. People supposed to follow a process, but this one does it this way.

This one skips process, you know, segment two, this one, you know, stays too long on segment two, right? Ai, AI understands all of that context. Wow. So, and then just ease it up. And I, and it will matter. I mean, we're gonna get. Better traffic patterns, we're gonna get better. Um, the communication and switches are all gonna get better.

'cause AI is just going to sit there and Yeah. Never get bored. You know, it's, it's like the termin error and never get bored. It never stops and never, so it's just gonna say, okay, I got all this data, here's what you need to do different. Yeah. It's a pattern, pattern recognition. Yeah. Yeah. And it's amazing at that, right?

Um. Wow. Beautiful. So that's my love it. That's my thread. Okay. Question number three. Um, what does AI readiness mean to you? And then what would you say to someone just starting out play you guys? I mean, like, um, I, and I say it work and with friends and you know you, and it is. It is the, it is the, as an actor, the word play has, I just love to, to dig into the literary definition of play, right?

Mm-hmm. So you've got all of these, um, elements. A play is a thing that is an organized set of actions that are gonna be taken and play is something that children do. So, but in this case, it is. You know, more that if you want to get good at it, more that childlike exploration. Mm-hmm. And I, and I think that people that open themselves up to that are going to be ready.

Now there's practical considerations here, right? So, um, those. Thinking a little bit about this, you, you've gotta make sure you have access, right? I mean, so I mean, that seems silly, but if you wanna be ready, make sure you have access. Mm-hmm. Um, we have in our organization, made sure that everyone has access.

Um, you have to. Prepare yourself to provide context, right? Mm-hmm. This is an Excel spreadsheet. This is something different. And interacting with it requires context and you have to set yourself up to curate Yeah. Um, curation, whether it's a highly creative fund project or whether it's a tech, a highly technical one, or it's equally important to have curation skills.

Um, I don't know what the end of that is. I have the, the dystopian view of it is we're all just gonna become George Jetson and press the button once a day to keep it going. But I think, I think it's less dystopian that, I think that's where, um, the, the human AI partnership is really important because the human's ability to curate in an abstract way and in a productive way is, is really important.

Beautiful. Yeah. Beautiful. Well, HT this was absolutely awesome. Thank you. Nice. Thank you for jumping up and doing this. I know, I know. I reached out to you at the last minute and you're like, I'm in. Let's do it. So I really appreciate that. So I wanna, I wanna compliment you guys 'cause just really quick, the um, so, and this is good.

I set brand new chat PT agent at the podcast. 'cause to be fair, I knew a little bit about it, but not a lot. And I said. I instructed agent to go to YouTube, pull the transcript for every single podcast, give me a summary of what it's all about and let me know a little bit about the best way to perform in there.

And I just want to read you a couple snippets from this 'cause it did an amazing job. And I'll share, oh, this is so cool. Oh, cool. Let's hear it. Um, this is the last sentence and I kind of love it by understanding the podcast ethos. Curiosity, creativity, ethical awareness, community, and playfulness. You can tell your stories and insights to resonate with Kyle Shannon, and the AI Res Readiness Piled project audience.

Ah, Kyle Shannon and, and the AI Readiness Project. Audience, I need to. Blown up on my reading skills, but I love that. That's curiosity. Community and playfulness is what it told from the summary of transcripts. Oh, and there's more elements of this you guys would probably really like, but I will Yeah. Share it.

Share it with us. Maybe we'll turn that into marketing, marketing fodder. That's awesome. Yeah. Did a pretty good job. That's amazing. Pretty impressed. Good. Awesome. I mean, that really is the desired effect. Like those words are what? We want this to be. Yeah, well, exactly. Nailed it. So, exactly. Well, well, and so did you.

Thank you for Yeah, yeah, exactly. Thank you so much. Being a great guest. Yeah, this was awesome. This is awesome. And, uh, we'll, we'll wind things up here with the. Lovely title image that looks nothing like you. It looks like I'm in a Cohen Brothers film. I, it really does. This is what it looks like, uh, when, when the people that make your graphics don't prompt enough.

Go See you later.