DISCLAIMER: Today’s Advertiser helped make the production and delivery of this episode possible, for free, to you! But the themes expressed by the guests do not necessarily reflect the views of Inside Reproductive Health, nor of the Advertiser. The Advertiser does not have editorial control over the content of this episode, and the guest’s appearance is not an endorsement of the Advertiser.

Please Note: We recorded this episode two months prior to release, and Manish and Ravi have already been pinging me about changes that have happened since. I will do another episode with them because this topic is constantly evolving!

Chat GPT is here to change the future of your job in the fertility industry, or maybe even take it. Is this a stretch? Dr. Ravi Gada and Manish Chhadua discuss how Chat GPT and its predictive technologies has the potential to revolutionize is already revolutionizing the fertility space. And what may come next.

Tune in to hear:

Uses for Chat GPT in fertility clinics and the Open AI source behind it.
How Chat GPT is being used to share data with patients, aggregate data, how it may be used in the future to generate prompts and consult notes.
The elimination of scribes and schedulers.
How Chat GPT will be able to interface with patients to provide 27/7 availability and access to care.
Griffin push Manish and Dr. Gada about what the second and third order consequences will be from this development, and what significant long-term impact it could have on the future of REIs.

Dr. Ravi Gada:

LinkedIn: https://www.linkedin.com/in/ravi-gada-md-mba-a2307527/

Manish Chhadua:

LinkedIn: https://www.linkedin.com/in/mchhadua/
Website https://reuniterx.com/

Transcript

Dr. Ravi Gada 00:00

In the fertility space, what we're going to deal with is who owns the data inside the EMR. So, when we talk about regenerative AI and language modeling, we're talking about being able to talk back and forth with a patient, maybe summarize a chart, create a summary of a consultation and put a note in the EMR. But we also talked about in AI, this whole idea of helping predict outcomes for IVF, as well as dosing for medications for a cycle embryo growth and development and who owns that data.

Sponsor 00:31

This episode is brought to you by Univfy, email Dr. Yao at mylene.yao@univfy.com, or just click on the button in this podcast, email or web page for your free IVF artificial intelligence tips and strategies. Today's advertiser helped make the production and delivery of this episode possible for free to you. But the themes expressed by the guests do not necessarily reflect the views of Inside Reproductive Health, nor of the advertiser, the advertiser does not have editorial control over the content of this episode. And the guest's appearance is not an endorsement of the advertiser.

Email Dr. Yao

Griffin Jones 01:13

A monkey can do an IVF egg retrieval. That's something that more than one REI has said to me. That's a euphemism. That's not really true. But will we be saying that what Rei is can do today is like the intellectual capacity of monkeys, based on what's coming with artificial intelligence? That's the topic of today's episode, you might listen to today's episode and wonder is Griffin high? No, the topic of today's episode is exactly why I don't get high. I talked to Ravi Gada. Dr. Ravi Gada and Manish Chaddua. Both of Reunite RX niche is the founder. Dr. Gada is their medical director, Dr. Gada also practices at Dallas Fort Worth Fertility Associates, we talk about chat GPT, which many of you have heard of some of you may have not the open AI source behind it, we talk about the applications that it's having. In the greater context right now the applications that it's having in the REI practice, how it's being used to share data with patients, how it's being used to aggregate data, how it will be used in the future for prompts and generating consult notes, how it will replace the work of scribes and schedulers and nurses how it will be able to interface with patients as an avatar of you. Because of technology that already exists. Today, I pushed Dr. Gada and Manish To explain what they think the second and third order consequences will be from this and what the REI will do when half of the communication half of the tasks that they're responsible for today are done by artificial intelligence tomorrow, at least half depending on what length of time we're talking about. And if we're talking about a long enough period of time, does it become everything that an REI could possibly do in a way that they couldn't possibly add any more value over what general artificial intelligence can do? You'll notice throughout this conversation, we really tried to keep the conversation about the applications of what happens in the REI practice, at least for half of the episode. But there's almost no way to contain it to just that open AI is Chat GPT product is just the tip of the iceberg and it has implications for every single aspect of the human experience. I might sound dystopian or pessimistic when I'm trying to get Manish and Ravi to think about this during our conversation. I don't think I am I think I'm pretty neutral. I'm not making a value judgment if it's good, bad or neutral, but follow along as we discuss how this conversion of technology not only replaces workflow that happens in the REI practice, does it replace the concept of human production altogether. Buckle up. Don't even consider consuming anything that has cannabis in it and enjoy this conversation with Manish Chadduaand Dr. Ravi Gada. Dr. Gada, Ravi. Mr. Chaddua, Manish. Welcome to Inside reproductive health.

Dr. Ravi Gada 04:06

Good to be here.

Manish Chaddua 04:07

Nice to meet you.

Griffin Jones 04:09

Manish , do you know how many times Ravi has Monday morning quarterback my show and I get a text or an email something that I should have said or something I should have asked. I've always asked him to come on. He says no, I don't want to rock the boat. I don't want to shake salts. I don't want to stir the pot. And finally I got a text from a couple like a month or two ago saying okay, I got a topic let's talk about yet. GPT. And I said all right, great. This'll freak people out. And he said companies government I said Yeah, so I want to freak people out about chat GPT. But we were speculating before we even started recording how much of our audience knows what chat GPT is how many of them know about open AI the platform that it's built on? So why don't we start Elementary and just give context for what we're even talking about?

Manish Chaddua 05:00

I think a lot of people have read a handful of articles maybe about chat, GPT. But you know, it's an endeavor that kind of started probably about five years ago. It's often invested heavily into it. And then, you know, really just back in November of this year, last year, they basically launched this first kind of forward-facing view for consumers of what exactly it's capable of. And so the founders behind it are, you know, a handful of guys, Sam Altman, Peter Thiel, Elon Musk, I think are some of the original core for it. But since Microsoft has invested upwards of $10 billion into this product,

Dr. Ravi Gada 05:38

well, and Griff just, I don't know if people know what I mean, Sam Altman is the former CEO of Y Combinator, Peter Thiel, former founder of PayPal, Elon Musk, obviously everybody knows. So it's got some pretty big backing behind it.

Griffin Jones 05:54

People know those names but tell us about what Chat GPT is doing.

Dr. Ravi Gada 05:59

Chat GPT is an AI language modeling platform, it's probably considered a SASS platform where users can go onto the web, create a login, it's absolutely free to use, there is a paid version of it that you get a little bit more priority. And you can ask it just about anything. And it has over 100 billion different data points. But you can ask it, you can just talk to it. If you're like, Hey, how are you today and go through a conversation, you can ask it? What's the reason for having an Hmh of less than one, you can ask it to draft legal documents that you can ask it to write a poem. So and really, it puts this together and you can iterate on it back and forth to get to the point where you're happy with the answer, copy paste, but it into your platform and use it a lot of people are saying it will be used to augment the workforce and make our lives easier.

Griffin Jones 06:54

Manish, How does that work? Like how is Chet GPT using open AI to be able to do that?

07:01

So chat, GBT is called the term that's being used for it as generative AI. And so what chat, what they've done is they've basically created, you know, in the term is a caucus of data of about 170 billion data points, which is articles, publications, all sorts of data points across the internet, they stopped collecting that data in about 2021. And really, the way that it works is actually through algorithms and just math, it's predicting the probability of the next word or the next most likely word in how it's generating this text. And so that's kind of the clever thing about it is that it's this large, large data set, it's able to basically look at that data set, and then predict the profitability of the next word. And that's how it turns into the text that gets outputted when you're asking your questions and the context that it actually receives when you follow up with that question, and things like that. So it's a predictive model,

Griffin Jones 08:01

Doctor Gada, give some of the use cases that Chat GPT is being used for what are some of the funky ones that you've seen, one of the funky examples that I've seen was, like, talk about a certain type of story in the tone of comedian Tim Dylan, and it was the comedian, Tim Dylan reading it. And it was pretty close. And even he says, like, wow, this is, this is pretty close. And it clearly wasn't there yet. But it's more than just write a poem or write an article, you can actually say, write an article for this certain type of audience or write it in the spirit of x. And so what are some of the wacky examples that you've seen?

Dr. Ravi Gada 08:43

People are predicting this year, chat GPT, or any other language modeling system is going to write a screenplay for a movie, it's going to give it some input data on what type of movie at once and who the characters are, it's going to write the movie from start to finish finish. And they're going to take that storyline and put it into an animated AI platform dollies for pictures, but there's some animated ones in the background, and it'll create the animated movie and that by the end of this year, we'll have a movie in which the screenplay and the animated movie are all done by AI with minimal human input. Wow. So even

Griffin Jones 09:21

the characters, the action of the animation is going to be created by artificial intelligence.

Dr. Ravi Gada 09:27

Yep, completely based on the language of the screenplay, and it'll make all the action of the characters, the voiceover to voiceover as well. So you can there's voiceover you can do now, so I could probably record all of your podcasts, uploaded it to chat, GBT right what I want Grif to say and replay it, and it's going to sound like I'm doing a podcast with you. And we can call it something else.

09:48

Well, I'm going a step further from that they can actually model based off of a handful of pictures of you an actual animation of your face and have that talking as the actual animation for that. Voiceover so that's so they can mimic like real life people and things like that. And that's not just GBT. But that's other AI solutions that are out there.

Griffin Jones 10:08

Sure. What is that? Is that the deep fake? What is that?

10:12

It's related. I mean, it's in that vein. Yeah, exactly. Yeah.

Dr. Ravi Gada 10:15

Deep fakes, probably the most popular one.

Griffin Jones 10:17

Is that a different type of artificial intelligence? What's behind that?

10:23

Yes, sir. I don't know a whole lot about what's exactly behind that. But it is using AI to basically evaluate facial expressions and things like that, like deep fakes, specifically takes my facial expressions, and it superimposes your face onto it. There's other versions of that that basically will just take text and known kind of vernacular and how mouths moves and things like that, to basically create video or animations of a person actually talking.

Griffin Jones 10:51

Okay, well, I could just totally dive into this part where I'm deeply concerned about someone making a podcast episode.

10:59

That's a really weird edge case, or not weird, but just kind of scary, is that even hackers are using chat GBT to generate clickable content so that way, they can send email blasts out and they'll just ask it things like, hey, create a email that's basically has a link in it, that's the most likely to be clicked by users. And it'll actually generate and so this is another edge use case where it's like, okay, well, you know, the malware the ransomware type of folks out there using it to help move their cars.

Griffin Jones 11:32

Well, I want to come back to this and talk about what we think second and third order consequences might be of all this, but let's talk a bit since this is, after all, a show for Rei is it isn't Rogen were talking to fertility specialists and the people that own fertility practices? What are the applications that open AI can be used in the REI practice at this time?

Dr. Ravi Gada 11:59

So I think at large, right, we've, we've seen in our space companies that come in just using AI for data mining for embryos, look grading eggs, grading embryos, there's companies trying to predict what's the outcome of an IVF cycle. But we haven't really seen too much movement in the linguistics modeling or the language modeling. So in an REI practice, could you create a chatbot that basically communicates back and forth with a patient answering simple questions. So if a patient calls, or has a question about what's my Hmh level? Or what's this thyroid function test, could could a language model reply back and forth with that patient just enough to answer as many of their questions as possible? In healthcare, you want to be very careful in what we call follow up criteria. So if the if the bot doesn't know the answer, then say, Hey, let me get one of the nurses for you or one of the doctors and then someone picks up the conversation from there. But you could think about that in a way where patient has free access or 100% 24/7 access to a chatbot that's been trained by us in the REI community. We've given all the language the data points, the conversational pieces to have. So that's a use case. Interestingly, I did a did a thing the other day I put write a male male couples gestational carrier contract, and it spit out a gestational carrier contract immediately. And then I said, Well, can we add language for what happens in the first trimester if there's abnormal screening, postpartum does the gestational carrier provide lactation and milk for them and and it added all these sections in there along with by the way, an exhibit page to add the financial conditions of all of these things, so I can have it write contracts for third party reproduction pretty easily. I had a patient asked for a work excuse the other day, and I had chat GPT write a work excuse after an abdominal myomectomy for six weeks, and it wrote it for me. It leaves blank so you know template so then you copy paste it and then you add the patient's name, sign your sign it and send it.

Griffin Jones 14:12

Let's talk about the EMH level example for a moment, the thyroid function example for a moment, how would we know if the bot gives the wrong answer?

Dr. Ravi Gada 14:21

So this is the part that gets complicated, right? So what's interesting is there's plenty of companies today that have language modeling, ai, ai ai, so chat. GPT is owned by open AI, open AI is primarily going to become a Microsoft based company. Recently, Facebook launched one called llama and then Google launched one called Bart and so everyone's going to have a version of this. You have to then take their AI language modeling and input your own data set. So perhaps that's recording the next 1000 hours of calls with nurses and physicians with their patient. inputting that data. And then running tests to see is it doing what it's supposed to be doing? And if it is perfect, if it's not, you have to give feedback to the system always. And that's how it's why it's called machine learning or regenerative learning is it has to learn from itself. The patient either has to tell it, it's wrong, the nurse has to tell his strong, but you've got to feed that system enough to be smart enough to give the right answer and smart enough to know when not to give an answer. But that's going to be the biggest challenge in our field is making sure it doesn't overstep its bounds.

Griffin Jones 15:33

And so at what point do we expect it to be able to be a better judge than a human being?

Dr. Ravi Gada 15:41

I think in some cases, we might already be there in certain language modeling. I mean, when we in you open up your Gmail, or Outlook, and it practically finishes your sentence for you when you're typing up an email now, and sometimes I'm like, well, that's better than I would have wrote it. So let's just go with that. But in the healthcare space, I think we're I think we're a bit of ways I think we always are later adopters, for new technologies for that reason. But if I had to guess, I mean, we have to be three to five years from being able to really, I hope within three to five years, where they're where we can leverage this type of technology.

16:14

And the biggest challenge is going to be what Robbie's talking about this Fallout criteria. So when we think about AI, and basically, you know, creating the answers are basically predicting what the answer should be. The probably the, the hardest part is going to be that aspect of just knowing when not to answer and AI is not there yet, or doesn't seem like it's there, which is why a lot of stories are online about how they're tricking chat GPT and providing wrong answers to math questions or, you know, doing a handful of other things like that. So that's probably going to be where, you know, some, the physicians in general, will view this as a tool that helps them get to the answer faster. But it's still, you know, we're far away away from between us getting to the point where we can blindly trust that to do that.

Griffin Jones 17:06

Have you read anything about the regulatory bodies or the agencies thinking about how we're going to regulate this either from ama or from Fe cog or from is anybody talking about this? Rob?

Dr. Ravi Gada 17:19

I don't think anybody's talking about him. I was listening to a couple of podcasts about it. So in healthcare, it's not interestingly Moniece mentioned to me earlier today, chat, GBT did certify that their HIPAA that they have a HIPAA compliant API version to it. I don't think any of the society organizations are talking about it. Even in this sense of copyright. People haven't really quite figured out when chat GPT pulls language from the internet, essentially rewrites that language and spits out an answer. It's not giving attribution for where that came from. And so there's even concern that could chat GPT ultimately get stuck in lawsuits with copyright? And are they just rewriting someone else's language or or verbiage that's out on the internet without site citation of credit? And Google does it right you type a search? It gives you an answer. But there's a link to where it goes from they might summarize a little bit in the in the description part. But ultimately, it gives credit through a link which chat GBD does not. So there's some people looking at this, but I mean, no society organizations from a medical standpoint, no, I don't think anybody is even digested what this technology means

Griffin Jones 18:31

until they hear this podcast. And they're like, Oh, crap, we have to have a board meeting.

18:36

And one of the counter arguments to the copyright thing that Ravi just brought up is that, you know, do humans in general do anything different? Are we just basically absorbing information and data from a variety of sources, and then basically mimicking what we hear with some amount of, you know, how much innovation is actually being produced? Out of what we regurgitate? Right? Some attribution

Griffin Jones 18:59

and some innovation, but very often isn't even possible to totally attribute everything because like the machine, you might be saying, in this case, money's we're aggregating and it's an amalgam of everything that we've consumed. But I was I was going to ask you that question about intellectual property, too. And you brought up the example of Google Ravi. And I wonder if if case law is still been established about that? Because sometimes I think like, is that enough when a creator is putting out information or creating something, and Google just kind of takes it and they put it on a Google search? And yeah, they give it a little bit of credit, but very often, what does the Creator actually get from that credit? If that person gets their answer right there in the search, they don't ever have to go to the creators website. They might see that little URL at the bottom, but they're pretty much just getting their answer from Google. Is there any kind of case law that, you know, Manish that has been established? Or is there are there battles going on about this

20:07

definitely is something that's been brought up even just about how the way Google works. Now Google gets a little bit of away with it, because they are actually providing that attribution. And I think that's where chat GPT will be very different. Because, you know, it's not the Texas generating is somewhat unique, but it's not actually sourcing that direct place of where the data is coming from. Even Ravi and I have had conversations about this as well, just to say, you know, here are the different differences. And then, you know, Google is a little bit different of an animal as well, because it's giving that attribution, it's giving hope to those creators to actually get the clicks or get the referrals. So I think it's a little bit of a different scenario altogether.

Dr. Ravi Gada 20:48

But there, but there is, there is case law for this. So there is something called fair use for copyright. So fair use has been established that our case law underneath that there's four criteria for violating fair use, but one of them is not citing the person, but it has to affect their ability to monetize. So if you have a company that has a bunch of articles about fertility, and you're regurgitating their data and putting it there, and their whole business model is to get links, have people click on that? And then ultimately buy something or lead them down something, then? Yes. And that's where Google pays people for that link. And so there is it's called fair use. I mean, I don't know that it applies to copyrights. Specifically, it's not going to apply to what we're going to deal with in the fertility space. I think in the fertility space, what we're going to deal with is who owns the data inside the EMR. So when we talk about regenerative AI and language modeling, we're talking about being able to talk back and forth with a patient, maybe summarize a chart, create a summary of a consultation, and put a note in the EMR. But we also talked about in AI, this whole idea of helping predict outcomes for IVF, as well as dosing for medications for a cycle, embryo growth and development. And who owns that data? Is that the patient is that the clinic? Is that the EMR? Is that everybody? And I think there will be a little bit of information that comes out from probably not the fertility space, but probably more on a higher level of internal medicine or diabetes of who owns this data.

Griffin Jones 22:27

I wonder if this affects people like me even more so than it might the general public and that those that are in deep niches, and are based around information are in deep niches, part of the reason why anybody picks a niche, whether it's a client services firm, or media company or software company is so that they're delivering specific needs to a small group. And that's where they that's the entire reason why they do it. And if something can just say, hey, take everything that inside reproductive health has gathered and created from original sources, then it could be the small niche companies that are most vulnerable, don't you think?

23:16

Yeah, I mean, content creation is something that's going to transform quite a bit. I mean, even if you look at the way, you know, traffic gets generated, and Google and even beings a algorithms work right. Content Creation is like, been the pinnacle of how they judge what's what's good, what's not what's new and fresh. And so that's definitely a large area of impact. I mean, there's, there's sub companies from chat GPT that have already been created that just create copy, and they create everything from sales, copy, marketing, copy, blog copy. So that's definitely distinct part of I wouldn't call it a threat, but a possible, you know, a rethink of that approach of copy creation or content creation.

Dr. Ravi Gada 23:59

I think the niche markets will get saved in this because when I look at health care, people focus on cancer, diabetes, hypertension, obesity, and fertility, and very small sectors get overlooked. And so all of these companies I think, are going to be focusing on the big three, you know, in terms of hypertension, diabetes, obesity, and then add cancer, and infertility kind of gets overlooked. I think that's why actually, as a field, I feel like we're very technology deficient. We don't have enough technology infused into this space, and maybe will be saved. I don't know.

Sponsor 24:35

Artificial Intelligence is revolutionizing the fertility space. Often we hear about imaging and automation AI in the embryology lab, but there's a different kind of application of AI and machine learning that is radically changing fertility patient care. One company uses AI and machine learning to analyze structured data. Structured data means the data sets in a fertility center’s EMR export their billing data and billing records. One company is using that data to get the smartest information to inform clinicians and patients in service of truly personalized fertility care. Univfy builds IVF prediction models for your specific fertility clinic, using your clinic's own data validated by your own outcomes. What are all the things you can do with a center-specific prediction model? How can you use it to increase patient satisfaction and IVF conversion? Dr. Mylene Yao, CEO of Univfy Fertility has tips and strategies for how to use AI and machine learning in your clinical practice. Email Dr. Yao at mylene.yao@univfy.com, or just click on the button in this podcast, email, or web page for your free IVF artificial intelligence tips and strategies.

Email Dr. Yao

Griffin Jones 25:57

We talkedabout a couple of the applications that you're using right now what applications do you expect that aren't quite there yet, but that open AI chat GPT will be able to do in less than three years.

Dr. Ravi Gada 26:11

Imagine a day that we're I'm sitting in a consultation room with a patient, there's a TV screen behind me here. And I say well, let's take a look at your Hmh level today. And on the screen, it hears me say that and pop to the h a m h up on the screen behind me for the patient to see that. And then I say, you know that's numbers normal, you know, that should mean that you have a good ovarian reserve. We also do a follicle count to look at that. And here's me say follicle count from your ultrasound. And it puts that up on the screen. And I have this now interactive conversation with the patient. They're asking me questions, we're going back and forth through a return visit or new visit. And at the end of that visit, we walked out of the room, I hit done on the recording device. And it generates the entire consultation note immediately on its own because it's regenerative language modeling gives me You know, I can then sit at my desk take 30 seconds to read it finalize it done, by the way, any edits that I make to that note that I didn't like the way it writes, it's recognizing that I edited it and and learning from that. So I think at the highest level, you could look at that you could look at it basically, you know, every six months, every three months, it reads the entire chart for a patient and summarizes it in a note on a three month update or a weekly update depending on what cadence you want to do that in. So there's things like that there's things that I could have it recording all the calls that my nurses do to patients, right, I rely very heavily on my nurses to communicate back and forth with patients. And I can and the language model can tell me if there was inaccuracies being presented or something that is different than what I would have said based on its understanding of the conversation. And then we can we can retrain that nurse, we can improve things, you know, it goes beyond nursing, to imagine the day that all of these things are just used as tools to make us better, more efficient. And ultimately, it will probably take over half of the I wouldn't say conversation but communication that we have back and forth with our patients.

Griffin Jones 28:24

At what point might we expect to see the avatar Doctor Gada, having that follow up. And so if all of those things are just different data points, and it can compare it to all of the data points from every piece of scientific literature, fertility and sterility is ever covered. And everything from all of the medical colleges, if it can just deliver that type of information, and we can use your video we can use your voice at what point are patients just seeing a virtual Doctor gotta

29:00

so I think the humanity and US will fight that pretty pretty well. So I think if you look at telemedicine, a lot of things like that, I still think the preferences is face to face communication, I don't think you can replace that for some people. Right. And I think for places where we're underserved pay at places where we're trying to get into that aren't getting quite the availability of health care. I think those are the areas where you'll see this kind of really explode or really thrive is to care for patients in in those particular areas.

Dr. Ravi Gada 29:32

But I mean, I've talked to Manish about this you know we have a lot of pilot projects in this area of where where will this technology take us and how do we get in a lot of it's in the datasets that are fed into the system but when I do think does the day come that you asked the patient would you like to see the human doctor or would you like to see the avatar Doctor initially or virtual care models are already there today. Many patients are going online and wanting to order their her own tests and get their information at home or through virtual care. So I think there's a version of it today, I think there's going to be a more sophisticated version of it in the future.

Griffin Jones 30:10

I'm a little skeptical on Manisha is hope for the humanity aspect. I think people want the humanity when they feel that the robot is insufficient. So the reason I yell into my phone when I'm talking to the the banking teleprompter is because it doesn't understand that I'm saying, talk to an agent or review account balance. But if I actually could do that as easily as I could correspond with a human being, I think it has more to do with convenience than humanity.

30:42

Yeah, for sure. And grip, I think my my point of view on that is more for general, for healthcare, I do think fertility is a little bit different, because of the age of the patient and kind of, you know, the fact that every fertility patient coming through as either a for the most part is Millennials or younger, right? You definitely could see this avenue of I'd rather text with my doctor than, you know, talk on the phone with them, or, you know, have to go and show up at a clinic and actually have that face to face interaction. So I definitely could see that scenario.

Dr. Ravi Gada 31:13

You think about this, there's a YouTube video out there, if you type robotic reenact the Moses of bow, using artificial intelligence, there is a cadaver. So it's a pig model of a robot, taking bow and sewing it back together without any human doing it and it healed intact. And then obviously, they checked it sacrifice the cadaver and checked it. And so, I mean, if we're getting to the point where cars can drive themselves, robots can do bowel reenact the most surgery on their own, we will get to the point where communication back and forth with the patient or consumer will get there. The question is how far right do you get to the point where you just do the intake form? And asking a few questions for clarification? Or do you deliver lab results deliver? Do you deliver positive and negative pregnancy tests and that way? That's the part is how far will it take it? I think it's going to go. If you fast forward 30 years from now, there's going to be a way different version of doing this. The question is in the next three to five years, or while we're all around, how far are we going to get?

32:17

And that's absolutely right. Like you take any technology, any innovation like this, and it's all a matter of a timeline, you assess some rate of improvement, and every tech pundit will say that is whatever the rate of improvement you select, that means at some point in time, you know, the technology will surpass the reality.

Griffin Jones 32:36

m&e, as you said, this has been in the works for some time now the technology behind chat GPT. But it seems like there has been an inflection point recently though, no, like, just how good chat GPT is itself. And then I practice with it. And a couple other like, think of translate for exam I, I don't remember the last time I used Google Translate for language, but it used to suck and not too long ago. And recently, I when we were covering the KKR story for buying ie vrma. And their only media coverage was in Spanish. And I speak Spanish pretty well. I put it into Google Translate to see and it was good. I like almost as good as as a native speaker who had been natively raised in both languages. So what's the inflection point when he's what happened recently?

33:28

Yeah, so this is common, right? This is common in a lot of technology, whether it's the smartphone or the internet, or, you know, even AI. And really, it's a byproduct of technology from 1015 20, even 30 years ago, becoming more accessible, less expensive to use, and basically more awareness, right? So you take smartphones from, you know, back in the late 90s. And they existed, and they had a lot of functionality. But it wasn't until the advent of the iPhone, where it really was the right time in place. And the cost equation made the most sense to where it can actually rapidly grow inside of that. And by the way, my background is telecom. So that's why the analogies there. But then pass that chat. GBT really is the first very consumer facing version of an AI model that showed the rest of the world everybody, including, you know, guys, like you and me, as well as you know, just college students and everybody else in between, right, what the capabilities of AI is. And I do think that AI has been in place for a long time. I mean, it wasn't, it was a number of years ago when AI beat, you know, IBM Watson mini in a game of chess. And this is just that acceleration. And I do think in AI, right, if we look at any of these revolutions that have happened, or major disruptions in technology, you know, it keeps happening faster and faster. And so So I think chat GBT has really opened everybody's eyes to what's capable? And now, all the thinkers and innovators are out there? Basically saying, Oh, I didn't realize we were this far along. How can we employ this as a part of, you know, a core model? Or how do we adopt this and find out what the right solution is that's really chasing this already, and integrated into our workflow.

Dr. Ravi Gada 35:18

And Griff real quickly to add on that. So the inflection point was I don't know if sometimes we will realize Chat GBT launched in November of 2022. So the inflection point was the first real launch of a major language model. And it obviously caught fire. And that's why we're all talking about it, or a lot of people are talking about it, interestingly, in that, but it was founded, I think, in 2019, four years, something like that about four years ago. And they've been working on it up until now, interestingly, post chat GPT launched, let's call it circa November of last year 2022. That put a lot of pressure on Google and Facebook to launch their versions. And so Google launched Bart, and they did a commercial about this. And in the commercial, Google asked, or someone asked the chat bot, to tell them about the James Webb telescope. And it was listing some bullet points. And the last bullet point said that the James Webb telescope was the first telescope to take a picture outside of our solar system, which was actually false, it was actually not yet planet and people picked onto that. And as soon as it did, Google's actual market cap value dropped by $100 billion that day, attributed to this error, because everybody said, their language model and their regenerative AI is not as good as Microsoft's, and they're not ready yet. And it lost some around seven to 10%. Market cap $100 billion because of that, but I think chat GPT launching in November is why we're at that flexion point today,

Griffin Jones 36:52

to the point that is a can take over half of communication that's currently happening between the REI practice and patients right now, maybe more than half so when that happens, Rafi not if because it will happen. It's only a question of time when that happens, what is the RBIs role going to be?

Dr. Ravi Gada 37:12

And you know, I mean, I think people worry about this a lot, right? People talking about not just the role of the RBI, but the workforce is these are these technologies going to replace the workforce. I mean, whether it was the calculator, whether it was Microsoft Word, whether it was, you know, all these different technologies that keep making us better and better. But we talk about this all the time in our field, that there's a under underserved population, there's, you know, we're at the tip of the iceberg. Maybe we're only meeting five 10% of the populations need. Does this actually make us better? Ultimately, we're still proceduralist we still do a lot of procedures in surgical procedures, egg retrievals, embryo transfers, IUI. Guys, so I hope or I think this is not going to replace the average ra i think it's going to make us more efficient. I think it's going to make our nurses more efficient embryologist more efficient. But you're right. How does it allow for us? And we talked about how many are the amount of retrievals that an REI can do in a year. And beyond that point, there. It's it's not beneficial maybe to the patient or the ER, and it depends how many nurse practitioners do you have underneath you? How many nurses? Well, this is going to be another adjunct to that technology have an honestly a checks and balance. I mean, imagine the day where we have going into an IVF cycle. And I'm going to do for the physicians and nurses that listen to the podcast, a Lupron trigger. Well, there's certain things for Lupron triggers that you want to know you want to know that that patient has regular menstrual cycles and that they have a normal FSH level. And so the second you order a Lupron trigger, that the that the AI actually scours the EMR and actually pings you and says, Hey, I don't see an FSH level on this patient. Are you sure you want to order a Lupron sugar? And I say, Oh, I'm glad it caught that. Let me order a FSH level real quick and make sure. So I think it'll make us more efficient. It's, you know, replacing us I think we're all going to be replaced one day, you know, whatever, whatever, you know, sector you're in, you're gonna get replaced 100 years ago, everybody was a farmer, or at least knew somebody was a farmer. Today, I don't really know that everybody can say I have a first degree relative. That's a farmer. So machines have already replaced, farmers machines have replaced manufacturing jobs. And that's the worry about this type of AI technology. It will replace jobs, but it will also create jobs. I mean, we didn't have the jobs we have today that, you know, that didn't exist 100 years ago. In fact, I don't know what the population of the US was 100 years ago. Let's make it 100 million people. Today were 300 million people, no manufacturing jobs, very few farming jobs, and everybody's still employed. So there will be new jobs created. Maybe we'll figure out newer ways to help people get pregnant, but things that are replaceable at Everybody should be looking at saying, you know, how do we either make ourselves better to stay ahead of it? Or how do we use it to, you know, augment what we do today?

40:09

And there's there's a lot of people out there far smarter than us that have kind of pondered upon this question as well. One of the other things that I think is kind of changed recently, is initially they thought a lot of low skilled labor would get replaced fairly quickly by automation and AI and things like that. I think chat GPT tests that a little bit and saying, Hey, listen, well, you know, if your job is sitting behind a desk at a computer, basically, replying the emails and doing things like that, there's a lot of risks there, probably more so than a surgeon, or, you know, even a mechanic at that point in time. So I think that's what it's changed kind of some of the view of what would get replaced by AI first, but I do think we're still a fairly long ways away from that, like, years, at least,

Griffin Jones 40:56

well, for now, and I do want to talk more about that. And we'll definitely end on a note where we're really freaking people out, but, Robbie, I want you to think a little bit about what it is that the REI will be needing to do in these coming years as Chat GPT gets an AI in general gets more sophisticated, like how I'm envisioning it is there's human Gada overseeing a hunt the capacity that robot Gada can do and robot Gada is helping to treat 100 patients and human Gada just needs to oversee robot Gada or is that not the right way of thinking about it? Because the human will soon not be?

41:38

Grip? I think the jury's still out on whether or not Robbie's a robot or not.

Dr. Ravi Gada 41:43

It could be it could be, do you wanna see dr ga da, or Dr GA D Ay ay ay.

Griffin Jones 41:50

Oh, it's already there. And and so what's the relationship supposed to be? Yeah,

Dr. Ravi Gada 41:56

I mean, I think ultimately, that relationship kind of goes back to, you know, we already use or have our staff help us accomplish what we accomplished in the day, I don't accomplish in a day, you know, very much if I don't have a nurse, an embryologist, a medical assistant, a billing person. And this will do the same. I think that, but I do think you know, we've managed to have talked about there, I'd love to do a commercial where I have four consultation rooms running with a iPad in there that's actually has my own avatar, speaking back and forth with that patient, one patient, it's their new patient console, the second room is their return visit with their lab results. And the third patient is coming back for another FET after a successful delivery. And all the while I'm actually over in the operating room doing the retrievals all day. I mean, so that day is coming. Now the question is, is that coming tomorrow? No. Is that coming in the next three to five years? Probably not? Is that something that we can work towards in the horizon of a 10 year type cycle? I think so. I mean, I know that might not sit well with some people. But I think you have to embrace this technology. We are looking at this very heavily. We're investing a fair amount of resources to figure out how to do that. And I think that the people that do will do well, I think the people that resist it may do well. But I think there's a high chance that they're not going to be able to be as efficient if they don't adapt to technology, which is the story over the last 100 years.

Griffin Jones 43:30

You talked a bit about it's some of this like data entry type of work that is most vulnerable. And I was hearing one expert on this topic talk about that it's actually more white collar work that is vulnerable rather than blue collar work because blue collar work tends to be more manual. But Manish when are we going to see an intersection between robotics and this type of AI because once that happens, then we don't need human God at all, once we have a robot that can do the very sensitive maneuvering in surgery that the best surgeons can do right now. And we have the artificial intelligence of all of the data points gathered from every surgery ever electronically recorded. When can we what progress are we seeing towards robotics and artificial intelligence? converging?

44:31

You know, it's actually something that's, that's familiar, before even AI right, it's the separation between engineering and technology or software. Right. And so this is I think, why we're seeing this is because replacing things that are soft like on a computer or something like that becomes a lot easier once you can get over a kind of the intellect or the brain of it, right? The biggest issue with robotics right now is probably the expense and so when In the cost of robotic arms, robotic equipment and stuff like that, that's reliable and high precision and things like that start coming further and further down. That's when you'll see this kind of cannibalize even those types of industries. And so that's where I feel like, you know, this low skilled or blue collar laborers, you said it, you know, as a little bit more protected, because the cost of those robots has not come down. And the functions that they pervert perform, and the accuracy of what they do, just isn't quite as inexpensive as, you know, your email solution of being able to message back and forth with patience or something to that regard. So it's going to happen, but it's just, you know,

Griffin Jones 45:42

so maybe there's a silver lining to all of this supply chain crap that it's slowing down the inevitable

Dr. Ravi Gada 45:49

grip. I don't know. Are you old enough to remember the Jetsons? I mean, that's where Yeah, remember

Griffin Jones 45:53

the Jetsons Flintstones crossover?

Dr. Ravi Gada 45:56

Yeah. So you know, I mean, imagine I mean, the Jetsons is looking forward to, obviously, if robots robots replace what we do, and we work, everybody's concerned on what would we what maybe we start enjoying life again, you know, we worked so hard, we, you know, is a society. And I'm not talking about just fertility, I mean, globally. And maybe we actually, you know, a 40 hour workweek becomes a 20 hour workweek. And we actually are able to read and spend time with family and travel. And maybe I mean, robots taking over and doing certain things. I'm not saying they're taking over the world. But maybe we get back to the point where society actually has time to do the things we do rather than being in this hamster wheel that we are in today.

Griffin Jones 46:38

Before it does, what other applications do you see elsewhere in the fertility industry and quote, so you talked about the applications that can happen in the practice between fertility providers and patients? But where can what other applications are we seeing right now with open AI, if any, in the fertility industry, and what more should we expect?

Dr. Ravi Gada 47:01

Yes, I don't think we're seeing I mean, I haven't seen it, I tried to keep a pretty good pulse on what's going on. I haven't seen it. There's some chat bots that are out there. But overall, in terms of chat, GPT, I don't think so we've seen it in obviously, in the lab, there's a lot of work being done to robotics and, and automation and AI. But what's interesting is, I don't, I think also no one in the fertility space, or even a lot of other spaces are going to actually be able to build their own technology on this, they're going to have to leverage I mean, think about Microsoft, Google, Facebook, Amazon, few other companies, I'm probably leaving out, but they have the best of the best, the brightest or the brightest, and essentially unlimited budgets relative to ours to do this. So a lot of this is going to be creating API Interfaces into their technologies. And using our datasets. I wouldn't be surprised if the EMRs that are out there are looking at this today, right? The electronic medical records, they're fairly technology forward, they are probably looking at their datasets, because they have actionable datasets. You asked me hey, you know, Hey, Ravi How much does DFW fertility associates? What kind of data do you guys have to feed into Chad GPT. And I've looked at you and say, I haven't even I don't have data. Like, I haven't started gathering that. But maybe I should, maybe we should start recording every conversation we have in the office with a patient and with each other, myself and my nurses, myself and the embryologist to feed this dataset, and is one individual, clinic or user or even an MSO going to be able to create enough data, perhaps but but likely not, it's going to require a collective effort amongst the industry. So I don't think we're there in terms of that. I mean, like I said, there's the earlier stuff, I was telling you writing a letter writing a contract for third party reproduction. But in terms of the high level stuff, it's got to be a concerted effort of gathering that data, putting it in, and then really, ultimately, you know, garbage in, garbage out. So if you put garbage data in, you're gonna get garbage data out is what that term is. But you've got to do that, then you've got to test the model over and over and over again, because in healthcare, we demand 99% excellence, right? In other industries, they might say 80% lunch, this, you know, we've all talked to a, a answering machine bot on a customer service line, they'll get to 80% and be satisfied with the quality of that work. We have to exceed that above 99%. So no one's there yet, but the question will be how do we get there? I think that a lot of people like us and others are looking at this. And I think that it's around the corner. If you ask me what does around the corner mean? I can't tell you the answer.

Griffin Jones 49:54

So I was going through Dr. Rudy Giuliani's workflow with her and I How she did 1300 retrievals last year and I was thinking of each of the points, she was talking about listening well, I could impact that I could impact and I told her, I said, You should listen to this episode that I'm going to record with Ravi and Monique, because she was talking about her scribes. And I was just thinking your scribes are gone, man, they're not they're not going to have a job in a couple of years. There's no way in schedulers to right.

Dr. Ravi Gada 50:23

Yeah, yeah, exactly. Or are their job changes, right? You know, they, you know, they either they're either gone, you're correct, or it changes, right. So we still like concierge service, right? So they, the bot kind of does that. I mean, Google right now, I think has a platform that you can order a pizza now through a bot or make reservations at a restaurant. And it'll actually if the restaurant doesn't have something like open table that you can go online and do it will call the restaurant and make the reservations for you and interact with the hostess without, without a person, it's a robot talking to a hostess. So those jobs will be either replaced or used in a different way.

Griffin Jones 51:03

Sometimes those applications come and they circumvent solutions that you would think need to happen, right? So for one of the things that we've been saying for many years is that millennials don't want to talk on the phone. But Gen Z absolutely won't talk on the phone. So you guys have to figure out your scheduling, you got to figure out this digital scheduling as well. Maybe you don't, because this Gen Z person can just input into chat GPT called the fertility clinic and make an appointment for me.

Dr. Ravi Gada 51:34

Yeah, that'd be ironic, as we keep focusing on how can we get the clinic to be the Chatbot. And we find out that the Gen Z is actually or the chat bots, and we're still interacting with them on the human side? Well, unfortunately,

51:45

they're not gonna go to the metaverse to schedule appointment anymore. So

Griffin Jones 51:50

well, it's just kind of one of these principles that you think of that we often it's, we have to build a certain type of infrastructure. And there were many countries, for example, that never really built out a telephonic infrastructure never had landlines at scale. And that was probably in their government central plan that, okay, 10 years from now, we're going to build telephone poles and have the wires out to the rural countryside. And they just never had to do that. And so there can be a number of applications that we're thinking of, for artificial intelligence that just circumvent the need for us to build out some other kind of solution.

Dr. Ravi Gada 52:31

So the other day I took I had an Excel sheet, it was a financial Excel sheet. And I took it and I was just curious, because I had heard people were doing this, I copy and pasted it, I didn't format it. And I thought what happened, so I just copy pasted into chat GPT, it looked awful. And I hit submit. And it summarized the Excel sheet for me without even having cells or columns or anything, it was very oddly formatted. So imagine taking the entire data set that we have for IVF patients and outcomes, and just dumping it into this thing. And just at first go saying, What do you think of this? Or tell us in patients less than with a Hmh? A 42 year old with an AMA H of 1.2? Whose BMI is this? Who has unexplained infertility? What what what what should we do? I don't know if that will be the answer that we're looking for today. But that's what we're probably looking to strive for. And, and that's literally just copy and pasting an Excel sheet. Imagine once you get these API's start working with these companies, and you really integrate with them to provide this type of data. I think it's, I think it's also like people, it freaks people out. But I think that when literally, when the calculator was invented, people thought, no one is going to know how to do math, we're all going to be stupid, nobody is going to use their brain anymore. And they're just going to rely on this device. And here we are today doing way, way more amazing things and advancing technology. And the calculator is a tool that you just use, and honestly half of us have moved away from that to things that are on our computer now.

Griffin Jones 54:15

Okay, so we can spend the next 10 to 15 minutes concluding this topic with going down these rabbit holes, because this is going to be fun, what you just brought up Ravi, the example of the calculator, how it's going to make people dumb, and people aren't going to know how to math do math anymore. Ravi, that did happen for probably 80% of the population. They can't do math anymore. And May and 20% can do math into levels of application that we had never even anticipated before. And probably a square root of that number is, you know, has just magnified the Einsteins of the world. But isn't that number getting smaller and smaller and smaller. smaller and the, the applications are greater and greater and greater. But eventually doesn't that number just become nil, because there's nothing that a human being can do to add value to collective general artificial intelligence,

55:17

definitely the edge of what we're talking about, I think Robbie talks about, like these alternative purposes for humans, and basically, what's going to create our, you know, Will and an ability to keep driving forward and stuff like that. And I do think that that those things will happen. But I do think there's a lot of fear around just that, which is, hey, listen, does the population as a whole get less intelligent? Or does a proportion of the population become less intelligent, and then you have this, you know, small niche of the population that continues down the road of research, and basically innovation and stuff like that. And that, you know, that's entirely the storyline of that time machine movie. So so i think i digress to the point

Dr. Ravi Gada 56:02

where it is, right. I mean, people have, maybe, maybe people have become worse at math worse at spelling, because Microsoft Word and everything auto corrects your spelling. And the older generations, like, gosh, we knew how to do all this, I feel like that sometimes. But the newer generation says, Well, you might know how to do math, and you might know how to spell. But these influencers are able to create a whole new, you know, industry, and they're able to create content, videos, edit it with through a computer that does it all with them. And it would take me eons of time to do that. And they can do that in a matter of an hour. And it would take me days, and I still might not get it right. So I might know what you know, the square root of 256 is and they're like, well, that doesn't matter. I've got a computer to help me do that. But you can't use the computer the way I can. So smartness is dependent on the tools that we have, I think that it, it forces people to be resourceful, and be able to use the tools you have. So just like you use a calculator, just like you use Microsoft Word, you're gonna have to learn how to use AI, and whether it's chatting GPT, or some other platform. And someone else might say, well, I could have written a beaut, I can write a beautiful act or essay on my own. Well, that's great. But if someone else can use a tool to do it 10 times faster and 10 times cheaper, they're probably going to win the race.

57:32

And we've seen this from a software point of view, we've seen this over the last, however, long, 40 years or so, right? Where software is now becoming easier and easier to produce, even what developers can accomplish in just a day versus what we had to do to do you know, back 20 years ago, just to get the same type of thing done has has totally changed. And so there's a rate limiter at some point in time where it's not going to matter that they can do more faster, because there's just not more to do. But we're not there yet, either. So, you know, our developers use chat GBT already today and just in the last few months, right? It helps them solve problems faster, it helps them optimize code that code faster, and a lot of things like that. But we have a long way to go before it replaces any of the developers. So

Dr. Ravi Gada 58:19

by the way, for for like normal people speak that like language model. This thing can code because code is a language so it can actually code software. And people are estimating 10 to 20% of software at at big companies is already being written by platforms like Chen GPT.

Griffin Jones 58:36

I see what you guys are saying human intelligence, resourcefulness, resilience, that's only one category of concern that I have. Let's pause it for a moment that we remain committed to innovation that we use this time, Robbie, like he says the possibility to be free to pursue other creative pursuits to enjoy life. Let's pause it for a moment that we don't actually get worse at anything. There still comes a point right? Where there is nothing that human intelligence and creativity can do to surpass that which a general artificial intelligence can think of let's let's think of ancient hominids, for example. It's some point they were equal at some point, humans parted with chimpanzees, and they parted ways with other previous hominids. But then not we live in a world where there is nothing that a chimpanzee can do to add value in a human being world other than be observed and be a pet. So doesn't that happen at some point? Where Yeah, no,

Dr. Ravi Gada 59:36

I mean, it's a great point. So what's interesting though, remember, AI and regenerative learning is data. Data input. So right now, someone estimated chat, GBT has 190 billion data inputs and it regurgitates it out. But it doesn't know what to put out unless it's been put in. So Chad GPT, for example, is likely or any AI is is likely not to figure out How to create this nuclear fusion between protons to generate energy, human intellect still is able to do that, right? They call it the neural network inside of AI. And what's in there is what's been inputted by humans. So a lot of people are saying that what's inside of the datasets, there'll be able to, you know, AI will be able to find it faster, regurgitate it, remodel it continue to do that. But it's always going to need to use or I say always, I should say, as of today is it needs source data, it needs innovation. So innovation is still going to come from humans. And we're going to do that. And then we submit it into a platform such as AI, and go from there. But as of today, I don't know that anybody has any great use cases of AI solving a problem that humans needed to invent or get to, it's really regurgitating all the things we have. And it's just gathering it faster and spitting it out faster. Maybe one day, we'll be able to have, you know, its own neural network that actually generates new ideas, but new ideas are still created by humans and put into the computer software system.

1:01:12

So I do think that there's some places where we're getting there, right. And that has to do with the sheer sheer compute power, right? This ability for it to go after large, large sets of data, right, and basically go through every permutation, right? So it's a little bit different from what we would think about as like new ideas. It's not necessarily a new idea. It's just a, hey, we've gone through every permutation of possible outcomes. And that's how we get there. And so there's, there's this, you know, looming threat or looming kind of, you know, fear of the fact that hey, listen, there's not anything more that we can do that hasn't been done by AI. But I do think that's right now, it's science fiction, at some point in time, it probably will become reality. But hopefully, it will be past my time.

Griffin Jones 1:02:02

The operative phrase that Dr. Gaga was using was as of today, and I think it's okay, as of today. But even Manish can think of a couple of applications where it's starting. And so what about what how long is as of today lasts for? Is it 10 years? Maybe? Is it 100 years? Probably not? Is it 1000 years? Almost certainly not? Almost? Certainly not?

1:02:26

Yeah, in grip. The interesting thing about that is that it's not a conversation about RBIs at all right? No, it's, you know, it's a

Griffin Jones 1:02:33

human race. Yeah. But it's the relevance of the human race.

1:02:37

Yeah. But even before that just passed, are you guys it's, you know, a cure for fertility, right. It's basically, you know, what's the pursuit? What's the purpose for, you know, humans and its happiness, and, you know, procreation and all these other kinds of facets. And so yeah, we'll get to a cure to fertility probably sooner than unnecessary need for humans.

Griffin Jones 1:03:02

I actually think it's going to be the thing that puts us all out of business, because I think it could even it could happen before a cure for fertility. I've said this for years that my long ball sci fi outcome is that,

1:03:16

but it'll be sustaining, right? It's putting us all out of jobs in order to sustain us otherwise, even the AI has no purpose without humans, but

Dr. Ravi Gada 1:03:25

it puts us out of business for what like we all are doing things so that we can be productive and earn money and then use that and enjoy life and have a purpose. But purpose will be redefined as it just as it was 100 years ago, where it is today. And it will be redefined again and another 100 years.

Griffin Jones 1:03:44

So I actually think it puts us out of the business of production. I mean, the the intersection of artificial intelligence and of virtual reality, I think that's what ultimately puts us out of, of the business of human production. Because when we can live in a world where we can augment our intelligence with artificial intelligence, so human beings are already cyborgs. This these devices that we carry around on us help to us to augment our intelligence and our communication abilities and all of our memory and then once that becomes further integrated with our brains with our nervous systems, and there's a virtual world in which we're able to participate, then eventually, what do you even need to reproduce physically in this physical world for you can have your augmented intelligence baby in your augmented reality world that never has to worry about dying that never has to worry about sickness that doesn't have to worry about human suffering. And I'm not saying this to you guys are smiling. Most people are going to be listening to this episode and not watching it so they can't see you smiling right now. I'm not saying this to be dystopian. I think this is just what's actually going to happen.

Dr. Ravi Gada 1:05:00

about maybe it puts us out of the business of being productive production, but it actually puts us back into the business of relationships and, and, and leisure and lifestyle.

1:05:10

And, and just to just to touch a little bit on the philosophical side of this, right, is just keep in mind the lifespan of a human is part of evolution. So,

Dr. Ravi Gada 1:05:24

that was pretty deep. I don't even know what that means.

Griffin Jones 1:05:26

Yeah, explain that many.

1:05:29

Yeah, so just kind of getting to the point that like, humans live the span of life that they live as a part of how we've evolved to become where we are right now, there's plenty of animals that live many, many years longer than humans and plenty of animals that live much shorter years than humans. And so, you know, that's, that's part of the equation as well. And, and the second thing that's kind of goes into that is it like, listen, we might have purpose with AI, but AI has no purpose without humans, either. Because what does a bunch of bots running around, servicing themselves and doing things for themselves me, either, that's a, that's a purposeless kind of function in that vein as well,

Griffin Jones 1:06:13

maybe, but I'm not convinced of that, they may find a purpose because the purpose of any living organism is just to continue existing. And human beings might be the first one to evolve itself out of existence. You talked about our relationship to other species in terms of how long we've been aren't, we haven't been around very long. It's been 200,000 years, I think, since humans separated from the last hominids. And when you look at our, our growth, it's been it's, it's a hockey stick, compared to the first years of leaving the canopy. And now civilization just in the past couple 1000 years, industrialization 200 years ago. And so I don't think this stuff is too far away. And I'm not trying to be dystopian, I just, I just don't think that I don't think there's any way for us to be able to contain it and control it. And so far you guys ever given otherwise?

Dr. Ravi Gada 1:07:09

You know, I think that people thought that when assembly lines came about, I think that they thought that when tractors came along, I think that is always been a worry. And it will always continue to be a worry. But ultimately, in a philosophical sense, humans are resilient. And like I said, we seem to stay ahead of the technology that we create ourselves. You know, at what point do we are we not able to stay ahead of it? Well, up until today, we still have I mean, people thought the world was over when assembly lines came in, and manufacturing jobs just got crushed, and what are we going to do and farming got replaced by equipment. And here we are today, three times the population with you know, 2% 3% unemployment, I mean, people are still employed doing something?

Griffin Jones 1:07:56

Well, if they said that, in the 1860s, as folks, were moving from steam to coal, you know, the late 1860s, or somewhere before the early 1880s. Whenever that happened, if they said, This is the end of humanity in the in the next five years, yeah, they would have been wrong. I think it's the amount of time where people get things wrong. I don't know if this is going to happen in a century or in a, or in an eon or a millennia. But I think it's inevitable that it will,