Latest / The Ask Mike Reinold Show / Using ChatGPT and AI in Physical Therapy

Transcript

Using ChatGPT and AI in Physical Therapy · The Ask Mike Reinold Show

• 0:50
  SPEAKER_02: On this episode of the Ask My Rhinal Show, we talk about using chat GPT in AI and physical therapy.
• 0:57
  SPEAKER_01: The Ask My Rhinal Show.
• 1:02
  SPEAKER_01: Helping people feel better, move better, perform better.
• 1:14
  SPEAKER_02: Before we get to the podcast, I wanted to make sure you knew about my free online course on the introduction to performance therapy and training.
• 1:22
  SPEAKER_02: If you want to learn how to get started optimizing and enhancing performance, this is the course for you.
• 1:27
  SPEAKER_02: Head to micrenault.com slash performance to sign up today.
• 1:31
  SPEAKER_02: Welcome back, everybody, the latest episode of The Ask Mike Round Show.
• 1:33
  SPEAKER_02: We are here up in Boston, champion PT performance.
• 1:36
  SPEAKER_02: Our crew, PTs, some strength coaches here discussing anything you guys want to talk about.
• 1:41
  SPEAKER_02: What questions do you have for us?
• 1:42
  SPEAKER_02: Head to micronaw.com, click on that podcast link and ask away.
• 1:46
  SPEAKER_02: Sports performance, PT, career advice, anything you want to talk about, ask away.
• 1:51
  SPEAKER_02: Still getting questions all these years later.
• 1:53
  SPEAKER_02: So keep them coming and we will keep answering.
• 1:56
  SPEAKER_02: So let me see.
• 1:56
  SPEAKER_02: Who do we have today?
• 1:57
  SPEAKER_02: While Lenny is frantically plugging in his headphones so we can introduce the students again.
• 2:02
  SPEAKER_02: I'll go slowly.
• 2:03
  SPEAKER_02: We have DeWesh Podell, Dave Tilly.
• 2:08
  SPEAKER_02: I'm trying to time this with Lenny getting his mic on.
• 2:11
  SPEAKER_02: Anthony Vedetto, Brendan Gates, Kevin Coughlin, Dan Pope, Lenny McCreena, and Lenny, who do we have for students today?
• 2:19
  SPEAKER_02: We're back.
• 2:20
  SPEAKER_02: Wow, that actually sounds good.
• 2:22
  SPEAKER_02: Len, who do we have?
• 2:24
  SPEAKER_03: Old school.
• 2:25
  SPEAKER_03: I love it.
• 2:26
  SPEAKER_03: What do we have for students today?
• 2:27
  SPEAKER_03: We got uh two lovely students.
• 2:29
  SPEAKER_03: Um we have Meredith Fagan from New York Institute of Technology and the Brian Santos from Duke University Blue Devils in Durham, North Carolina.
• 2:40
  SPEAKER_03: Nice welcome, welcome everyone.
• 2:43
  SPEAKER_02: Nice, everybody is welcome.
• 2:44
  SPEAKER_02: Brian, what do we have for a question today?
• 2:47
  SPEAKER_00: So Sean from North Carolina is asking, hey champion, I have a question regarding the use of AI as a physical therapist.
• 2:54
  SPEAKER_00: I'm a couple years out of physical therapy school, and I feel like I missed the boat on using AI while I was in school.
• 3:00
  SPEAKER_00: It's hard to keep up with it as it seems to keep evolving every month.
• 3:04
  SPEAKER_00: How are you guys using AI in your current practice?
• 3:06
  SPEAKER_00: Are there any specific apps or websites that you recommend and why?
• 3:10
  SPEAKER_02: Awesome.
• 3:11
  SPEAKER_02: Thank you, Brian.
• 3:12
  SPEAKER_02: And I think it's kind of cool that Sean from North Carolina was read by Brian from North Carolina.
• 3:18
  SPEAKER_02: There's like a connection.
• 3:19
  SPEAKER_02: I feel like YouTube bonded right there.
• 3:21
  SPEAKER_02: So really cool.
• 3:22
  SPEAKER_02: But um, I'm gonna start this episode and say that this is probably gonna be the fastest outdated episode we've ever recorded of all time, right?
• 3:30
  SPEAKER_02: By the time you hear this, or by the time you're looking at this in three months, we're probably gonna have different opinions because AI right now is bonkers in terms of how fast it's progressing, what's new every day.
• 3:42
  SPEAKER_02: Every model is accelerating at such a rapid pace, it's crazy.
• 3:47
  SPEAKER_02: So, whatever you think you're doing now is gonna be probably different in in three months, but maybe we can talk some like general terms and stuff like that.
• 3:54
  SPEAKER_02: So, you know, Sean, first off, I hear your point on missing the boat.
• 3:58
  SPEAKER_02: I think everybody on this podcast did not grow up using AI, right?
• 4:03
  SPEAKER_02: Like we didn't use it in high school, we didn't play around with it trying to figure out how to cheat on our homework.
• 4:07
  SPEAKER_02: So we're probably not as good at it as some of the the younger generation, but the more you play with it, the better you get.
• 4:14
  SPEAKER_02: Now, let's break this episode into two things, guys.
• 4:17
  SPEAKER_02: Okay, and so I'm gonna throw it at you guys.
• 4:19
  SPEAKER_02: One, what are we using for AI models, right?
• 4:22
  SPEAKER_02: Because there's Chat GPT, there's some others, and I'd like to hear some thoughts.
• 4:26
  SPEAKER_02: I know most of us are using them fairly often now.
• 4:29
  SPEAKER_02: But two is what AI apps are we using?
• 4:33
  SPEAKER_02: Because I we we actually have a few questions on this.
• 4:35
  SPEAKER_02: I tried to like kind of put them together, but what AI apps?
• 4:38
  SPEAKER_02: I know there's some running apps, we know there's some baseball apps, but what apps using AI have we tried?
• 4:43
  SPEAKER_02: So who would like to start?
• 4:45
  SPEAKER_02: Let me see.
• 4:46
  SPEAKER_02: Anyone jump in?
• 4:47
  SPEAKER_02: Anthony, you want to jump in?
• 4:48
  SPEAKER_02: What have you been trying there?
• 4:50
  SPEAKER_02: Anthony's probably are you the youngest of us technically?
• 4:53
  SPEAKER_02: Yeah, I think you could say that technically, yeah.
• 4:55
  SPEAKER_02: But like, correct me if I'm wrong, you you missed the boat on AI, technically, right?
• 4:59
  SPEAKER_05: Yeah, definitely wasn't around even just like three years ago when I was in school.
• 5:03
  SPEAKER_05: So I had no background for like, oh, I forgot to do a homework assignment.
• 5:07
  SPEAKER_05: Can like ChatGPT or like write it up for me real quick?
• 5:10
  SPEAKER_05: So I missed that boat too.
• 5:11
  SPEAKER_05: Um recently, I think most of us have come across something called open evidence, which has been an awesome resource and platform.
• 5:20
  SPEAKER_05: I think for myself, just to quickly um search.
• 5:24
  SPEAKER_05: Hey, I have a question about literature on return to hitting or whatever the case may be.
• 5:30
  SPEAKER_05: You plug that in, open evidence, and it gives you a list of really great resources, articles that kind of shows what might answer your question.
• 5:38
  SPEAKER_05: Um, click on those links, it takes you right to the article.
• 5:40
  SPEAKER_05: So that's something that I've been using quite frequently.
• 5:43
  SPEAKER_05: If I have a question about something in the literature and I don't really want to do a full lit review or lit search, um, so that's something that I've been using quite frequently.
• 5:53
  SPEAKER_02: Yeah, I completely agree.
• 5:55
  SPEAKER_02: I think open evidence is one of the rare ones that I trust.
• 5:59
  SPEAKER_02: And maybe that's gonna be my downfall, right?
• 6:02
  SPEAKER_02: But I actually feel like the hallucinations just aren't quite there.
• 6:06
  SPEAKER_02: But what you could do in open evidence is really cool is you can you can just ask a clinical question.
• 6:10
  SPEAKER_02: It's like, what's the evidence behind, you know, you know, lasers for wound care, for example.
• 6:15
  SPEAKER_02: I've been looking into that.
• 6:16
  SPEAKER_02: What's the evidence behind, you know, special tests to differentiate lateral epicondylitis, right?
• 6:21
  SPEAKER_02: You don't have to be a prompt engineer, which is like a new field coming up where you have to like make a fancy prompt, right?
• 6:27
  SPEAKER_02: You can just ask a clinical question and it nails it.
• 6:30
  SPEAKER_02: And more and most importantly, because I usually put the same searches in multiple like chatbots just to see the difference, it gives you references that have links and you go right to the PubMed article and you read it.
• 6:43
  SPEAKER_02: And man, open evidence is is number one so far for me.
• 6:47
  SPEAKER_02: So I I agree, Anthony.
• 6:48
  SPEAKER_02: Uh, Brendan, did you have one you wanted to talk about?
• 6:52
  SPEAKER_04: Yeah, I mean, I like open evidence a lot.
• 6:54
  SPEAKER_04: Uh, and I'm with you.
• 6:55
  SPEAKER_04: I think my favorite thing about it is it just lists all those references.
• 6:59
  SPEAKER_04: So, you know, you can go back and you can check those papers um just to make sure that it's not hallucinating.
• 7:05
  SPEAKER_04: So I like that a lot.
• 7:06
  SPEAKER_04: Um, I don't know about you guys.
• 7:08
  SPEAKER_04: I use ChatGPT a little bit, mostly for like fixing stuff at my house.
• 7:11
  SPEAKER_04: Um, but open evidence has definitely been the one that uh I use most clinically.
• 7:15
  SPEAKER_04: Um I wanted to share a little story that I thought was appropriate for this.
• 7:20
  SPEAKER_04: Um a couple years ago, uh I had an opportunity to present at the APTA uh Massachusetts annual conference, and the keynote speaker was actually an ED doctor from uh I think BMC.
• 7:31
  SPEAKER_04: And his whole talk was about AI's place in healthcare.
• 7:34
  SPEAKER_04: And uh going into that talk, I didn't think that there was going to be as much use case for AI in you know our field versus something like finance.
• 7:42
  SPEAKER_04: Um but he proved me like very wrong very quickly.
• 7:45
  SPEAKER_04: And uh he gave us a story about how they've been using kind of their own like proprietary chat GPT model.
• 7:53
  SPEAKER_04: Um and the standard example he gave was how to use it as a tool.
• 7:58
  SPEAKER_04: So um he told us a story about how a patient came to the ED.
• 8:02
  SPEAKER_04: Um, he evaluated him, he ran tests on him, he made a diagnosis, and his words were you know, long story short, he got the diagnosis wrong.
• 8:11
  SPEAKER_04: Um unfortunately the patient passed away.
• 8:14
  SPEAKER_04: Um and then after the fact, he took all of his objective data, his subjective uh interview, and he put it into his Chat GPT or their ChatGPT version and said to come up with a diagnosis.
• 8:26
  SPEAKER_04: And that chat GPT actually spit out the correct diagnosis.
• 8:30
  SPEAKER_04: I thought that was pretty staggering.
• 8:32
  SPEAKER_04: Um you know, he said, would the outcome have been different?
• 8:35
  SPEAKER_04: Like would the person still be alive?
• 8:37
  SPEAKER_04: Probably not.
• 8:38
  SPEAKER_04: He was pretty sick, but maybe, right?
• 8:40
  SPEAKER_04: And uh, I think it was interesting to hear that he said on that day, you know, Chat GPT or their equivalent was better at his job of diagnosing a patient than he was.
• 8:49
  SPEAKER_04: So um he told the rest of the crowd that now when the residents and everybody else do their rounds, they actually have two screens now.
• 8:57
  SPEAKER_04: So they have one screen with their charts so they can chart review their patients for the day.
• 9:01
  SPEAKER_04: Uh and then their other screen is this chat GPT model.
• 9:04
  SPEAKER_04: And so they'll use it as a tool to plug in uh basically all of their assessment for their evaluation, and then they can almost double check their thinking and bounce it off of them as if it's like a colleague, right?
• 9:15
  SPEAKER_04: Who just has you know access to a lot more info a lot quicker.
• 9:18
  SPEAKER_04: Um I I think where people in healthcare can use it and be successful is maybe you know, it's not gonna replace us, like it won't replace a sound assessment because it can only output what you input.
• 9:31
  SPEAKER_04: So you need to be able to give it good information for it to help you.
• 9:34
  SPEAKER_04: Um but I think the clinicians that can use you know open evidence or things like that as a tool and use it properly, it's probably gonna be pretty helpful.
• 9:44
  SPEAKER_02: Yeah, and I think that's such a great rationale, Brendan, why we should all be embracing this and giving it a shot, right?
• 9:51
  SPEAKER_02: And I could tell you from even my experience, I just started using it like with evaluations.
• 9:55
  SPEAKER_02: A new evaluation comes in, I try to prep it with a but a bunch of stuff.
• 9:58
  SPEAKER_02: And I'll be honest with you, as you get older in our medical professions, there's so much bias in our heads, wouldn't that just experiences that we've we've been through that like sometimes you like lead yourself to a conclusion because your brain wanted to, right?
• 10:14
  SPEAKER_02: Versus like having an open mind and being a little bit more thorough.
• 10:17
  SPEAKER_02: And I've definitely had a couple cases where it's like, oh man, geez, I I don't think I would have looked for A and B.
• 10:22
  SPEAKER_02: I just would have looked for C and D.
• 10:24
  SPEAKER_02: And you know, that that kind of opened me up to not just being stuck in my patterns of trying to find that.
• 10:29
  SPEAKER_02: So yeah, I agree.
• 10:30
  SPEAKER_02: That's uh that's that's pretty good.
• 10:32
  SPEAKER_02: Dave, what do you got?
• 10:33
  SPEAKER_06: Yeah, so off of gates, there's a really interesting uh story here.
• 10:36
  SPEAKER_06: So one of the big uh model stability AI was founded by a guy called EMAD, and he was a hedge fund manager who did very well in the finance world and then essentially was kind of just looking for things to do.
• 10:47
  SPEAKER_06: And it came to be that his son was diagnosed with a very rare form of autism when he was born.
• 10:52
  SPEAKER_06: And so he, you know, had the resources and opportunities to pretty much find the best doctors and all this kind of stuff.
• 10:57
  SPEAKER_06: And essentially, when he went to these different neurologists and doctors, they said that like his form of autism is very rare and hard to treat, and that there is like an ungodly amount of evidence coming out every day on that field of autism, that it's really hard for neurologists to read and process and see what's real, what's not the method.
• 11:14
  SPEAKER_06: So I think about a thousand papers on autism coming out maybe every few months.
• 11:18
  SPEAKER_06: It's hard to keep up with that.
• 11:19
  SPEAKER_06: So he pretty much, you know, said, okay, well, I'm gonna build an AI model to help process and understand research very well.
• 11:25
  SPEAKER_06: And he essentially did that and he built stability AI off the back of helping doctors better diagnose his son with this rare form of autism.
• 11:33
  SPEAKER_06: And the doctors were essentially saying that the ability to Gates' point to have my own brain, my clinical judgment, but have something analyze 100 to 200 articles for me in 10 to 15 minutes while I'm doing something else, and then I can go back and use my clinical judgment or whatever.
• 11:48
  SPEAKER_06: Um, that helped his son find a very like successful treatment uh method for his type of autism, which, you know, the neurologists were still there, everyone was still there, but they can't read 100 to 200 articles per day.
• 12:00
  SPEAKER_06: And so on the aligns of open evidence, I think that's where it kind of maybe summarize a lot of our thoughts is we have our clinical judgment.
• 12:05
  SPEAKER_06: And I I think the opposite of my brain is always like AI can't be empathetic, AI can't probe better questions, AI can't be, you know, when someone's starting to tear up in an eval because they're going through something really hard.
• 12:15
  SPEAKER_06: They can't be there for that, but they can definitely be there.
• 12:18
  SPEAKER_06: Those models can help us process, you know, somebody walks in with something you haven't seen in five to six months, and it's like, you know, give me the current evidence on 18-year-old hip label repairs with acetabular, whatever this fracture is that I haven't seen in six to eight months before.
• 12:31
  SPEAKER_06: That's where I think I'm finding the most use cases now.
• 12:34
  SPEAKER_06: Is it still goes through the lens and filter of me, but it helps me process a volume information that's impossible for me to do on my own.
• 12:41
  SPEAKER_02: And call me crazy, Dave.
• 12:42
  SPEAKER_02: When you wake up in the morning and you're preparing for the day, I know we all kind of do similar.
• 12:46
  SPEAKER_02: Like I'm I've been preparing for my day all morning, right?
• 12:48
  SPEAKER_02: Like me jumping in on open evidence and looking up some stuff, like because I get a new patient coming in and I'm looking up for recent stuff, it's refreshing.
• 12:56
  SPEAKER_02: I feel like I love it.
• 12:58
  SPEAKER_02: Like I feel like, like, okay, I'm I feel good about myself, feel confident.
• 13:02
  SPEAKER_02: I'm like, I'm I learned a couple things this morning.
• 13:04
  SPEAKER_02: Like, you know, and I'm old, right?
• 13:06
  SPEAKER_02: I'm like, you know, like it's it's you know, to me, I just it's exciting to be able to like be able to do that.
• 13:11
  SPEAKER_02: And before it was daunting.
• 13:12
  SPEAKER_02: That's what everybody says when we do episodes on like how to stay current, right?
• 13:15
  SPEAKER_02: Everybody says, like, it's too much.
• 13:17
  SPEAKER_02: I don't know, there's too much.
• 13:18
  SPEAKER_02: And they're right, there's too much crap getting published in so many journals and stuff like that.
• 13:21
  SPEAKER_02: Like, you don't know what to read.
• 13:23
  SPEAKER_02: So um, yeah, that's awesome.
• 13:24
  SPEAKER_02: Uh, Brendan, you want did you have something again?
• 13:26
  SPEAKER_04: Yeah, just it reminded me when Dave said the empathy piece.
• 13:30
  SPEAKER_04: Uh, this same keynote speaker that I was talking about, that was another part of his talk.
• 13:34
  SPEAKER_04: They actually ran some studies at BMC, and that that was a big concern.
• 13:38
  SPEAKER_04: Like, was the AI able to be empathetic?
• 13:40
  SPEAKER_04: And after their research, which I don't have, so I I can't prove it, but what he said was that the AI models were actually more empathetic than a lot of the doctors.
• 13:50
  SPEAKER_04: And uh, I thought that was crazy.
• 13:51
  SPEAKER_02: Um crazy and depressing and disappointing.
• 13:56
  SPEAKER_02: I don't even know.
• 13:56
  SPEAKER_02: It's like the most that should be the most empathetic field, but you just become like a a robot, I guess, you know, jeepers.
• 14:03
  SPEAKER_02: Uh Len, what do you got?
• 14:05
  SPEAKER_02: So, you know, as the senior, as the elder, right elder, you know, physical therapist without a doctorate degree, um, you know, just whatever.
• 14:12
  SPEAKER_02: Um, tell tell us about your experience with AI.
• 14:17
  SPEAKER_03: I use this this new concept called AOL.
• 14:20
  SPEAKER_03: Um, you you dial into this and now um AI with Jing Tut.
• 14:25
  SPEAKER_03: I I love everybody's responses, but I think we've kind of lost what the question was, which is some apps and websites that we're using.
• 14:32
  SPEAKER_03: So open evidence is a great one.
• 14:34
  SPEAKER_03: Yeah, but I think also, and I think you kind of touched upon it when you put things in different uh AI, um, you know, different AIs to test.
• 14:44
  SPEAKER_03: Um, I think for me personally, ChatGPT obviously, and I've I've subscribed to the uh$20 per month, which you know, another subscription because I think it's a little better, but it is a little crazy at times, so you've got to be careful.
• 14:57
  SPEAKER_03: Uh, I think also um Google Gemini is a really good one.
• 15:00
  SPEAKER_03: It has access to everything in Google.
• 15:02
  SPEAKER_03: It has YouTube, it has Google, all Google search, all Google information, anything that's in on Google, all the search engines, it has access to that information.
• 15:10
  SPEAKER_03: And I've also been playing with Grok, which is X, Twitter, whatever, Elon Musk's uh app as well.
• 15:15
  SPEAKER_03: And I find that one to be pretty reliable as well.
• 15:18
  SPEAKER_03: Um for some apps.
• 15:20
  SPEAKER_03: Um I use Swing Coach for um golf, maybe more me personally is a good one.
• 15:25
  SPEAKER_03: We were gonna talk about this.
• 15:26
  SPEAKER_02: That was perfect.
• 15:27
  SPEAKER_02: Yeah, exactly.
• 15:28
  SPEAKER_02: Like tell tell everybody a smidge about that too, because even as like a PT, like those similar things.
• 15:33
  SPEAKER_02: I don't want to break your thought, but like I want you, like that's important.
• 15:36
  SPEAKER_02: Like, like what does that app do?
• 15:37
  SPEAKER_02: And what do apps like that do?
• 15:39
  SPEAKER_03: So the app basically is is watching me move and it's detecting certain things that it knows is critical.
• 15:45
  SPEAKER_03: It's for example in the golf swing.
• 15:47
  SPEAKER_03: But I think there's also other ones for, and I think gates, I I don't want to put words in your mouth, but I feel like there's other ones for like um watching people deadlift or watching people squat and watching velocity of movement and stuff like that.
• 15:57
  SPEAKER_03: And I I don't remember the names off the top of my head, but those are valuable.
• 16:01
  SPEAKER_03: So it's watching you move, it's sensing where the club is in space, uh, compared towards, you know, maybe a pro golfer or the the elite golf swing.
• 16:09
  SPEAKER_03: And it's giving you feedback on like, you know, your golf, your club is too steep, your club is too shallow, and it gives you feedback to videos to help you um improve whatever.
• 16:19
  SPEAKER_03: For me, it'll be my golf swing.
• 16:20
  SPEAKER_03: So I think there's those concepts out there as well that people need to try to find.
• 16:24
  SPEAKER_03: Um, especially for us, like we said, you know, do we we use JimoWare for uh movement and velocity-based movement stuff, but there are apps out there.
• 16:32
  SPEAKER_03: You just put, you just record the video of them moving and it will tell you how fast they're moving.
• 16:37
  SPEAKER_03: You know, it'll give you a velocity-based kind of uh input or uh output of you know how how fast that was, and probably give you feedback on that too.
• 16:45
  SPEAKER_03: So I think again, there's a bunch of apps out there.
• 16:47
  SPEAKER_03: This is gonna be outdated in a few weeks or a few months, but I think right now that's what's kind of um you know, is I'm using and what is helping me with putting PowerPoints together with patient care um and also just kind of observing, you know, my myself and others and how they move.
• 17:02
  SPEAKER_03: So yeah.
• 17:03
  SPEAKER_03: Awesome.
• 17:04
  SPEAKER_02: Thanks, Len.
• 17:04
  SPEAKER_02: Pope, you got anything you want to add?
• 17:06
  SPEAKER_02: I know you're embracing it quite a bit.
• 17:07
  SPEAKER_02: And you know, I think we've covered a bunch, but what what's your what's your experience?
• 17:11
  SPEAKER_07: Yeah, I'm definitely afraid of AI taking over my career and my business.
• 17:15
  SPEAKER_07: So I'm trying my best to stay on top of it.
• 17:18
  SPEAKER_07: Um, I haven't looked into this.
• 17:19
  SPEAKER_07: I've heard this from several physicians, and I know you can do it for physical therapy, but uh having some apps that will listen to your evaluations and just write a soap note for you, like that's another no-brainer.
• 17:30
  SPEAKER_07: I haven't looked into this myself.
• 17:31
  SPEAKER_07: I think R note taking is very easy, so I haven't, but just me being lazy, right?
• 17:35
  SPEAKER_07: Uh the other thing I will say is that I use Chat GPT almost exclusively.
• 17:39
  SPEAKER_07: I haven't messed around with some of the other ones, but uh what I've found is that I was very frustrated with Chat GPT initially.
• 17:45
  SPEAKER_07: And what I learned over the course of time is A, Chat GPT is getting better, but I think it's important that you just start messing with it, meaning that you have to learn how to prompt appropriately.
• 17:54
  SPEAKER_07: So you can tell Chat GPT that you're a physical therapist, you work with patients, the only information I want is from PubMed or randomized controlled trials.
• 18:01
  SPEAKER_07: Um, how would I treat this specific person?
• 18:03
  SPEAKER_07: And if it knows all that information, it can give you something very similar to open evidence, or you can say, like, hey, I want the citations for every single thing you're saying.
• 18:11
  SPEAKER_07: And then if you don't know if one of the things it's saying is accurate, you can ask it, where'd you get it from?
• 18:16
  SPEAKER_07: Why do you think that?
• 18:16
  SPEAKER_07: How certain are you about this?
• 18:19
  SPEAKER_07: Right.
• 18:20
  SPEAKER_07: And what I will say is that I've I even open evidence will hallucinate or give you wrong information.
• 18:25
  SPEAKER_07: But the other thing is that, you know, I think older folks are getting um scammed by AI videos.
• 18:31
  SPEAKER_07: They're like, oh my gosh, do you see what Michael Jackson did?
• 18:33
  SPEAKER_07: Like, mom, Michael Jackson's been dead for, you know what I mean?
• 18:36
  SPEAKER_07: Just because you can now make those videos online and it's hard to tell if you've never looked at them.
• 18:40
  SPEAKER_07: But I think with AI, you have to interact with it and you can start to figure out where you're like, hmm, I think this is wrong.
• 18:46
  SPEAKER_07: And then you can dig a little bit deeper because that's one of the fears is that it's hallucinating, it's giving us wrong information, so we can't use it.
• 18:51
  SPEAKER_07: But I think that if you use it enough, you start to figure out where this seems like it's wrong, and then you can try to come up with uh the right um information by asking it closer or looking at the actual sources, so on and so forth.
• 19:02
  SPEAKER_07: So I think just try to embrace it.
• 19:04
  SPEAKER_07: Uh to me, it's like social media.
• 19:06
  SPEAKER_07: I guess I've been around around long enough to kind of go through that period and be frustrated by it because I didn't like it, didn't want to learn more about it.
• 19:13
  SPEAKER_07: But as you learn more and delve into it, I think you get better at it.
• 19:17
  SPEAKER_02: Awesome.
• 19:17
  SPEAKER_02: Thanks, Dan.
• 19:18
  SPEAKER_02: And I, you know, I would just wrap it up and summarize with a couple of points on on my thing.
• 19:22
  SPEAKER_02: Um, ChatGPT and Gemini have been going back and forth today.
• 19:26
  SPEAKER_02: Gemini is far ahead of Chat GPT, and it's crazy.
• 19:30
  SPEAKER_02: I don't think I would have said that uh last model for Gemini.
• 19:33
  SPEAKER_02: But Gemini right now is is fire.
• 19:35
  SPEAKER_02: Um, it's been so much better for me.
• 19:38
  SPEAKER_02: And I still put a lot of prompts in both just to compare.
• 19:41
  SPEAKER_02: Um, but Gemini is still winning right now.
• 19:43
  SPEAKER_02: So I'd say, you know, um, you know, maybe this is bad advice because you should you should probably go all in on one just to master one.
• 19:49
  SPEAKER_02: But gosh, they're they're essentially the same, like in terms of that.
• 19:52
  SPEAKER_02: But two other tips for you.
• 19:54
  SPEAKER_02: One is you can set up personal context in both of these.
• 19:56
  SPEAKER_02: So go into your settings and two things that I added.
• 19:59
  SPEAKER_02: One was um uh it's called the truth protocol.
• 20:02
  SPEAKER_02: I don't know what that is, but just Google it, you'll find it.
• 20:05
  SPEAKER_02: But I have this like very long thing in mind that says, like, you're acting under the truth particle, like, don't guess.
• 20:11
  SPEAKER_02: Give me verifiable sources, like, you know, never, you know, whatever.
• 20:15
  SPEAKER_02: Give me the percentage of you know how accurate you think this is, right?
• 20:18
  SPEAKER_02: And you put that in, and then every answer it gives you is based off that.
• 20:21
  SPEAKER_02: And then you can create like a custom gem in Gemini or GPT in in ChatGPT and just say, like, you know, I'm a sport physical therapist, you know, blah, blah, blah.
• 20:30
  SPEAKER_02: Like, only give me answers from verified sources like PubMed, like Dan said.
• 20:34
  SPEAKER_02: You can put those into your settings so that way you don't have to do it every time.
• 20:38
  SPEAKER_02: And then it just kind of feeds it out.
• 20:40
  SPEAKER_02: So that's what I would say to experiment with.
• 20:42
  SPEAKER_02: And I would say start slow in the morning.
• 20:43
  SPEAKER_02: You got some evalves coming in, type it in.
• 20:45
  SPEAKER_02: Say, hey, I got this person coming in.
• 20:47
  SPEAKER_02: What should I be looking for?
• 20:48
  SPEAKER_02: What should it do?
• 20:49
  SPEAKER_02: And I actually think it's like right now, AI is like trying to impress you.
• 20:53
  SPEAKER_02: It you'll ask it, like, hey, what are some special tests for later epicondylitis?
• 20:57
  SPEAKER_02: And it's gonna say, Yeah, but don't forget, also ask for these things in subjective.
• 21:00
  SPEAKER_02: And here's some red flags to look out for.
• 21:02
  SPEAKER_02: Like it goes above and beyond your answer.
• 21:04
  SPEAKER_02: So, you know, just try to embrace it and start.
• 21:07
  SPEAKER_02: And I think that's where you'll get going.
• 21:09
  SPEAKER_02: Okay.
• 21:10
  SPEAKER_02: So um, you know, hopefully that helped.
• 21:12
  SPEAKER_02: Uh, if you have questions like that, head to micronaut.com, ask away, and please subscribe.
• 21:16
  SPEAKER_02: We'll see you in the next episode.