Welcome to Real Talk on AI and I care with your host, Dr. Scott Morris. And I'm co host, Dr. Rayhan Ahmed.
In each episode, we discuss current news, adding a little AI education, and debate innovative AI tools and topics that will change the industry. Well, oh, everybody, welcome to Real Talk. This is your host, Dr. Scott Morris.
And I'm Dr. Rayhan Ahmed. So Rayhan, Phil, what's the news of the day today? Well, our news of the past week, you know, since something that doesn't actually come from I care or Silicon Valley, it comes from the US government, the Department of Labor.
So just a couple of weeks ago, end of March, the Department of Labor launched something called Make America AI Ready. And it's pretty cool. I was just using it earlier today. It's a text messaging education program to educate people like regular people on AI.
And you use it by texting the word ready to READY, READY, right? To to number 2022, 2020, and then over the next seven days, it will send you text message, it'll bite-sized educations about 10 minutes a day, and the whole idea is accessibility. Scott, this sounds a lot like what we do with the knowledge of AI. No, that sounds exactly like what we're trying to do.
I was like, exactly what we do. So I encourage people, this is AI 101. So it's really, it may be a bit basic for audience, but this is something that's great for your patients. It's great maybe for your staff or people or even for those listening who just want a little bit more.
I subscribed. I think it's kind of cool to see what other people are, you know, how people are learning about AI. It's one thing it's really interesting. Scott, it's all text based.
So you text a number, you know, text is number. It comes back and it asks you questions. How do you use AI? Do you have you ever used chat GPT or Gemini?
How often do you use it? And then it sort of so it's by directional education. They're also learning what we're doing too. Yeah.
I guess there's always a data motive, right? We talk a lot about that. But you know, my takeaways. This is not as we've been saying, you know, for past couple of years, the AI revolution is here.
I mean, this is like proof that this is, I mean, if you need anything more that there is, this is AI is real and it's going to be everywhere. So encourage our audience to take a look and definitely validates what we're doing. Scott in terms of our knowledge, right? So happy to present some, you know, maybe some good news from the government, a little bit sort of productive information.
So I thought those nice. Well, I just signed up. I'm looking forward. I got my message.
Thanks for your role in AI 101. You get your first lesson soon. That's pretty cool. I can't wait to see what they teach us.
All right. Well, so today, I have for the knowledge of speaking and already have for the knowledge of it. I have been dealing a lot with video generation this week. And so I thought maybe that was an interesting topic of, you know, how does AI bring pixels to life?
And so I mean, think about the last video you watch that it's core, you know, it's just a whole bunch of still images frames moving so fast our brain just sees it in fluid motion. But like if you sit there and you watch the Netflix and you stop, it's a still picture. Right. It's only when you push play that the next one and the next one and the next one and the next one.
We used to have to capture those frames of cameras or, you know, I think about Walt Disney, you know, is they draw those by hand. And now AI is kind of changing the game and, you know, and I also like, well, how does it do that? I think well, most modern videos using something called the diffusion model and starts by taking a whole bunch of random digital pictures and images. And like everything else we talk about, it trains itself on millions of videos to recognize certain patterns.
So when you give it a prompt, like, I don't know what we always talk about it, you know, a dog, well, dogs were in a spacesuit on, you know, in, in light of the moon mission, right. A dog's wearing a spacesuit on the moon. And it begins to strategically look through that digital noise and change those images into shapes and movement that things that look like we did look like we're used to. And so, you know, I was thinking about like generating we generate video for numerous of the marketing projects we work on generating a single image is hard.
I mean, it is hard to get it right. But then when you think of video, that's exponentially more challenging because like in a single image, if you take a picture of a hand, well, a hand just needs to look like a hand. It's just in one spot in a video that hand has to move to the fingers disappear. Do they reappear?
Do we have six fingers? We have four fingers with two thumbs. The lighting has to stay the same from one picture to, you know, from frame number one to frame number 600. And the AI doesn't actually understand physics or gravity.
It's just predicting what pixels should come next. So in some of these earlier models of video generation, we got some really interesting stuff that popped up and we're like, huh, okay, well, that wasn't quite what I was shooting for. You know, or you'll do it and then try to redo it and it comes out totally different. So these are called hallucinations and you know, and you've talked about those before.
You know, I was think despite the glitches and I still think we're early in the game. The potential for this is totally staggering. I mean, this is democratization of artistic talent. Like we've never seen.
You know, I just think about that by the young lady who works for me and does a lot of the editing for these real, real talks, you know, what used to take her weeks and weeks, if not months to generate of, you know, sitting there and pixel by pixel and changing things. Now she can do in three or four minutes and make it make the avatars talking any language. And our budget was well, whatever the cost of Gemini was that month. So, you know, and so now, you know, I think about that the creative ability that has a single person with a laptop that can allow for personalization, accessibility to video and images that maybe they would never have had access to.
I think that the ability to create a personalized customize. I think how is it going to relate to what we're doing. I think the ability to relate to build individualized customized standardized, if you will, like so we know the contents valid videos about education. I think that's right around the corner.
And, you know, I mean, it's not going to be without its pitfalls. I mean, first of all, video generation can sometimes give in misinformation. We often think about or that misinformation might be wrong data or it might be something we call deep fakes, you know, in the deep fake problem. Just continues to grow.
I mean, I don't know about you guys, but a lot of times now I wonder how much of tick-tock and Instagram and, you know, how much of that's actually real. I think we can still as humans perceive much of it and go, that was a weird glitch. That's probably not a real human. But, you know, give it two more years and I think we're going to have a tough time.
So I think there's concerns about, you know, copyright consent, where, where is the image generation learning from? Did it look at all the Disney films of Mickey Mouse and build them out? Well, somebody owns that data, you know, so, and maybe that data is not famous like Disney, but maybe it's a human artist who creates something really cool and put it on the internet and it's one of a kind. And now there's thousands of that same kind.
So I think there's lots of challenges, but I look at what we're doing with AI video and I think that I am not the most creative person in the world right on, but I think there is no limit to what the creativity of AI video is going to show us in the next couple of years. Yeah, I 100% agree. I think we are rapidly advancing to a future where maybe my kids instead of going on Netflix or to a movie theater, man, that seems to really almost pass a now, may just go on a prompt and say, you know, I'm kind of interested in something like spider man meets Iron Man in space. And you have a two hour feature like movie, you know, that that future seems very much within reach.
Oh, totally, totally, totally. So I think how does that, you know, maybe that's a good transition to exactly what we want to talk about today is, how is that type of stuff video generation or even text education like you did in our news. You know, how is that going to change how we as healthcare providers get educated, which I think we kind of covered two weeks ago in our real talk. But also how is it going to change how we educate and how our patients are educated with or without us.
Yeah, I think this is a topic we've touched on many times and we talk on for hours and hours in terms of marketing and outreach, I think as regards to education, I think there's huge, huge potential, I mean, just combine. I care with what the Department of Labor is doing and for your patients and that could be an amazing way to have a and doesn't have to be video, but it could be video of you or an avatar or whatever. To talk about common eye issues that come up in your voice or the way you want to treat it. So I think with your logo with your logo and having that by direction of communication, this is already happening.
I know, Braceton, I've involved with a couple startups that are doing exactly that. So this is already real and super exciting because this is one of those things where there's not sky and I we mentioned this before where there's not an obvious ROI. Like you're not bringing in you I would I would or you are, but you're saving a ton of time. You are saving us.
By the time this goes live, though we AI and I care in combination with many of the jobs and journals were kind of cross promoting a paper I just wrote called the real KPI revenue per minute. And when you think that the average history is five minutes and for most practices, the cost per minute of using a patient, depending on your practices somewhere between five and eight minutes. So it's cost in you $25 to sit there at a minimum to sit there and educate your patient about. Multifocal contacts or premium I well or glaucoma.
I think there is a huge ROI in that because what if I use that five minutes. Times 20 patients a day that's an extra 100 minutes of my time. How many more patients what I see in another 100 minutes five more. Yeah.
It's it's all it's all about demonstrating that I think to our fellow I care professionals on the immense amounts of time that maybe. That compounded over a day and a week and then a month it really starts adding up and that's one aspect. Then the other aspect is just delivering a higher standard of care to your patients. Can you imagine it's 10 pm at night and you know it's not an eye threatening issue.
But you just kind of question about a surgery that's coming up in two or three days and you just want to ask someone a question. It wouldn't be nice to just have your doctor available and if they're not available to the next best thing for something that could definitely wait. Well, these are ways to wrap it up, right? That explains the answer.
There it is. I think that's, I think that's massive in terms of, you know, people, patients, consumers, how everyone have referenced that, will actually have access to, as we said earlier, standardized education in a video format, like they're used to learning. I think about how we educate now, like we use in our office a lot of imagery and video to educate patients, but how many times do we go, okay, so you have a floater, if you think about, you have this jelly in the back of your eye, patients are going, will is it strawberry, does it see? Tell me, what do we do here?
You know, we kind of take that for granted because we've seen pictures or, you know, you've done surgery and vitrious is, you know, you're looking at, oh, there's vitrious, right? And so, so we in the eye care field go, okay, we understand what it looks like, but for a patient, they have no idea what that looks like. And so, they're having their own hallucinations about what it probably means, instead of being able to have a 3D live real animation of what it really means. How does that decrease the fear factor of a patient?
Yeah, exactly. We'd love to hear from those in our audience who are working on these solutions. So please reach out to us, Scott, I know many in our audience are interested in working on this sort of stuff. So I'd love to hear what's working and what's not and what's challenging.
Oh, we love to have you on the show. We love to have you on the show. If you're definitely working on this. And so please contact us.
My email, as always, is rayhan.ai and iCare.ai and Scott, you're Scott with one T at AI and iCare.ai. And of course, all our material is available, freely and accessible at a i and iCare.com. Looking forward to continuing the conversation. Have a good day, everybody.
Welcome to Real Talk on AI and iCare with your host, Dr. Scott Morris. And I'm co-host, Dr. Rayhan Ahmed.
In each episode, we discuss current news, add in a little AI education, and debate innovative AI tools and topics that will change the industry.