Would you trust it more if it wasn't called AI?

Automatic transcript of the recording (English); speaker labels are automatic. Timestamps open YouTube at that moment.

0:00Sacha: Welcome back to AI with Sach. Today, I'm thrilled to have Mike Tadasco on the show. Mike has spent years at the forefront of innovation from his time leading creative initiatives at PayPal to founding his own startup. He's an inventor with over a hundred US utility patents, holds an MBA from UC Berkeley, and has a knack for showing people how technology, and specifically AI, can reshape our everyday world. In a recent piece, he wrote, Mike posed a fascinating question. Would you trust it more if it wasn't called AI? And we'll dive deeper into that. First of all, Mike, it's such a pleasure to have you with me. Welcome.

0:43Mike Todasco: Sach, no thanks for having me. I'm glad to be here and excited to have a fun conversation with you.

0:48Sacha: So, as I've mentioned, you've had quite a journey from finance and accounting at GE to product and innovation at PayPal. Can you give us a bit of a snapshot of how you found your way into AI and generative technologies?

1:06Mike Todasco: Yeah. So, like, yeah, I mean, like, I started as a finance guy who was just really into technology. Now, that's like the honest answer. And, you know, I was just always into this stuff. Like, you know, 23 years ago, I actually had a robot mowing my lawn, an example, and stuff like that. Like, I was just very into tech, very into that. And, you know, I would just always be fiddling with stuff, even though that wasn't my day job. And really, it was kind of PayPal that opened that up. I did do an online startup before PayPal, just kind of figured a lot of things out through that that did not succeed. But when I got to PayPal, we were working on some really cool emerging technology and a lot in the world of machine learning and artificial intelligence. And I just got to get involved with that. The first product, the first, like, really emerging tech product that I was on was called PayPal Beacon, which does not exist. Most people have never heard of, but you could go on YouTube and you could watch the video for PayPal Beacon.

2:07And in 2013, it was a really cool technology. So I got to work on a lot of the emerging tech. And then comes 2016, when after PayPal has split from eBay and became its own company, I had a chance to run the innovation group for PayPal, which led to a lot of emerging technologies like blockchain and augmented reality and, and of course, AI and machine learning and all of these things. And I just loved it. I loved it every day. And, you know, the six or so years that was in that job. It was always super exciting. So it was just all of those things that kind of led up to it. And, and yeah, it was always weird to have a guy with a finance background kind of running the innovation group, but it worked out in the end.

2:56Sacha: And you found that your startup before the big wave of AI startups, how did your early entrepreneurial experiences shape your view on innovation and risk taking?

3:09Mike Todasco: Yeah. I mean, it made me think that I, I should have been building my startup on all the tools and services that I needed to build my startup versus the startup itself. Um, I was building an e-commerce marketplace and, you know, this was before Stripe and, and before like Braintree and all that. Like, so I remember when we started taking payments, cause I was just using people that I hired on, um, on Odesk at the time, which is now Upwork. Um, I didn't even have a secure checkout page or anything like that. Just obviously that's just table stakes now, but like these engineers were just building that. And like, like it was, it was just janky. Everything was so janky and it was so difficult like doing this. Like, yes, we were using AWS for storage. That was still relatively new in 2008 ish timeframe. Um, and I started working on that, but like just figuring out so many of these things. I I've always had a soft spot for small businesses and the tools that they need to succeed.

4:13And that's one of the reasons why, frankly, I joined PayPal after that, because it was building a lot of those services and giving them to companies so they can accept payments. But I think the thing like today, knowing that it's like, my gosh, if I were to create something else, like with all the AI tools that we have, just thinking like the, the PR, the social media stuff, the marketing, the legal, all of that kind of stuff that I was just doing and paying for out of pockets. I could be using Claude and ChatGPT and in some ways getting, you know, 90% of the benefits of actually paying somebody to do that and having it done in much less time and at a fraction of the costs when, you know, and when you're a startup, you don't have money to spend on those kinds of things. So I have a lot of sympathy for that. And, and frankly, I am most excited about the opportunity that exists for startups, for small businesses in the world of AI, because they've never had an opportunity before to have a machine learning engineer on site building stuff for them.

5:17Like they couldn't afford that. That was crazy. But now for 20 bucks a month, you have access to the best models in the world. And that's pretty amazing.

5:25Sacha: Yeah. Most definitely. I recently wrote an article called, would you trust it more if it wasn't called AI? And I think this question is so relevant today and it was, it was definitely relevant before, but even more so today. Your article focuses on how the name AI influences public perception and trust, right? What first sparked your idea to test this theory?

5:59Mike Todasco: I mean, I think it came from like in the 2010s when I was at PayPal, nobody muttered the word AI when they were talking about what we would now call AI stuff. Nobody said that. One time, me or somebody else like saying that and one of the machine learning, and it was all machine learning or deep learning or whatever the specific discipline is part of that. And I think it's because there were so many AI winters that had happened. And what an AI winter for folks who don't know, it's basically, there have been many cycles over the last 60 years or so where there's all of this hype. You know, if you go back to the 50s and 60s, there was a belief that some had that, wow, we're going to be able to solve AI, artificial intelligence within the next five to 10 years. I mean, like, and then that would happen and there's all this hype and so forth. They're like, oh, wait, we're not even close. And then a plunge and like these would happen and happen again in the 80s and so forth.

7:00So that's what the AI winters are is the period that happened after that. And I think for the people who were in that world in the 2010s, like you just didn't talk about AI. Like AI conjured images of science fiction and all this other kind of stuff. So it was just known as machine learning or whatever. And that's what I always called it until, I don't know when it was, but sometime like 2022, you know, probably before chat GPT and so forth when like it just was starting to get more into the populace and common belief and it just started to be called AI. AI. But if you look at, you know, when it was really coined AI, this was back at Dartmouth in the 1950s. There were a lot of other things that they were, you know, basically there's a whole bunch of people in this field, this emerging field. They didn't even really have a name for it. And they're like, well, what the heck do we want to call this thing? They were looking at all these other options like cybernetics and all these other things that they could have called it.

8:03But in many ways, AI was the most sci-fi. It was the most futuristic. And in the 50s were a time where, you know, the U.S. is battling with the Soviets. There's so much of a focus on science and, you know, advancement to science and eventually going to space and all that. Like, so artificial intelligence was the best for marketing. But where we're at today, I don't think it's the best descriptor of what the product is.

8:31Sacha: I love that. So you're arguing that calling something AI can make it sound a little more futuristic and but also even intimidating, right? Why? And I want to dig a little bit deeper into that answer that you just gave us. Why do you think AI has stuck despite, you know, these potential drawbacks that we're talking about?

8:56Mike Todasco: Yeah. Yeah. I mean, I think it's intimidating because it conjures up images of like a different separate entity from yourself. Like this is literally an artificial intelligence that we are creating. And it conjures well, like, well, you know, beings have intelligence. So, like, this must be its own being and so forth. When in reality, it's, you know, I almost hate using this phrase just as much in some ways, but it is really, really, really fancy autocomplete. It is what, you know, most large-flinguage models are now. And obviously, I say that because I don't want to be dismissive of them. Like, the models we use today are amazing, and you can do just incredible things from them, but still doing autocomplete. It's still not, it's not, there's not a consciousness underneath the hood or anything like that. And so, for most people who really haven't used generative tools, and that is the vast majority of the online population still who haven't used these tools, it conjures other images of them of, like, oh, my gosh, there's just this intelligence.

10:09It's going to take all of our creativity. It's going to take all of our jobs, take all of our this and this and this. And it might do that eventually. It ain't going to do that with what's out there now. That's for certain. And so, that's why I feel like there's a, probably a better, more rational thing to call it, because, like, it's there to augment us. And that's why I think in the piece I talked about augmented intelligence as part of it. Like, it is there, it's a tool for us. It is a tool to make us better in our jobs, just like a computer is, just like a phone is, just like whatever else, whatever other kind of tools we have out there. AI is just a tool for us to actually use out there in the world.

10:50Sacha: If you could go back and rename it, what term would you use? And how do you think rebranding AI helped reduce fear and probably encourage more, you know, widespread acceptance?

11:09Mike Todasco: Um, gosh, if I were to go back to the 1950s, it probably was the right thing to call it at the time. I mean, because, look, they're trying to get attention, they're trying to get funding, they're trying to get grants, all this stuff. They wanted to sound more futuristic. It's hard to fight against that. It had the most marketing punch of all of those titles. So, it's hard to say that they were wrong in that. It's the fact that it just stuck for 60 years. Um, that's kind of the hard thing, uh, with that. And, I mean, I'd probably just go back to, you know, today, let's call it augmented intelligence or something like that. So, I, I'm not a branding person. I'm sure somebody could come up with a better phrase than that. And I'm not the first person to use augmented intelligence, I'm sure. Um, but, like, that's more what it is. Like, these tools are invaluable in my workflow. But they, they are idiots when they try and run independently of anything I do. One of the things I've been doing since 2022 is I've been having an AI write a book, um, and then publish the book on Amazon.

12:18And I do this under the pen name Alex Irons. I give all the money to charity. Um, I think I've done it like 17 of these books now or something. And the books all suck. Like, they just do. They're no good. They're, they're just cliché. They're all this kind of stuff. And I'm always trying to use the mess model. I'm always trying to, like, but I let the AI go. And I just let it go and just say, hey, AI, tell me a story, make it amazing. And it can't do that. Um, will it eventually? Probably. It probably will get there at some point. So, like, um, but if I was doing this with the AI, then that becomes something different. Um, and through most of the writings I do, I will use AI, whether it's for spell and grammar checking, whether it's for researching, um, whether it's even just some brainstorming ideas if I'm stuck in a situation. Um, all of those things are great, great use cases to use AI. And it's ultimately there just to augment the things that we're doing.

13:19Sacha: I couldn't agree more. How do you, how do we ensure that, you know, AI remains a tool that serves humanity, as you mentioned, rather than, um, overshadowing it?

13:32Mike Todasco: Oh, um, such a, like, part of this, like, this is a really difficult question. It's a great question. And something we all need to be thinking about. Um, one of, you know, the most dystopian and utopian movies out there, I actually think is her, um, dystopian because the AI that he falls in love with, spoiler, spoiler alert here for anybody who hasn't seen the 10 year old movie, her, um, like leaves and just goes and does its own thing. Cause it gets bored with like, you know, the human beings that, you know, the AI was falling in love with as well. Um, but it's actually kind of utopian because when you look at everybody's job in the movie, it's, it's a super creative job. It's actually quite awesome. It is, you know, the, um, Joaquin Phoenix in the movie is writing love letters and he's verbally dictating them to the computer. Handwriting, doing the printing, doing all this other kind of stuff, but it is like literally a human being doing it. And, um, you know, all the characters in the movie are doing these really creative jobs.

14:38Um, but the AI in that world is just there to serve that better. How do we get to at least that part of the world? I don't know. I mean, I, I honestly don't know the answer because I think the one, look, the reason why I write all this stuff, because I do believe that the more common folk like me, frankly, work with and understand these models and the more that you use them, uh, the better off we're going to be. Like I live in Silicon Valley. We don't want like AI just to be dictated by a bunch of people in Silicon Valley who are like deciding the fate of humanity. Uh, the more that just people use these models, understand it, what they can do, what they can't do, how they'll help them with their lives. The better off those people are going to be. Um, and, and that's one thing I, I really do believe. We're going to probably get to the point at some point in the future where a model is going to be able to do everything that humans can do and do those things better.

15:42That point, what does the world look like? I, I, I don't know. It's not like all the jobs go away the next day. Um, there are many things where we actually prefer the imperfections of a human to the perfections of a compute. And I, and I would argue actually art and storytelling are going to be among those areas. So the, the creative arts, I think, you know, are actually going to be okay with that. Um, a lot of the other areas I, I'm just not sure about. And when we hit this point, it is going to be really disruptive for our society when we get there. Um, I don't know what the answer is. Is it a universal basic income? Is it, you know, banning all the machines out there or whatever? Like, like there's a lot of solutions. I don't know if any of these really work. Um, I think we've like, I'm sure, you know, there's a lot of bait. We, this is unprecedented in human times though. And that's the honest answer. I just like, it's not clear how this is going to land.

16:44But the one thing we can do today is at least just understand these models more and more. And the better that we kind of do that, the, the more benefits that are going to accrue to the workers, to the people that are actually able to then use these.

17:00Sacha: Yeah. Yeah. Most, most definitely. I do want to go back a little bit to your article. Um, you surveyed two groups on ethical views of, um, software versus AI. What surprised you the most in those findings?

17:19Mike Todasco: Yeah. So what I did for anybody, um, uh, again, I surveyed like 200 people and 100 saw a treatment A, 100 saw treatment B. Treatment A was something like, um, do you want to use AI to help you get from place to place with driving directions? And then treatment B saw something like, do you want to use Google maps to help you go get place to place from with driving directions? So basically it was just putting the word AI in where an AI already existed. Um, and this is just a very simple way in a survey. You know, if you do this against two similar populations of people, you're going to be able to basically test out, well, what is the impact of just the phrase AI versus something where AI is already happening? So, you know, most people don't think of Google maps as like an AI tool, but it is, it's not a generative AI tool, but it is an artificial intelligence.

18:21It's a, they're using deep learning and other techniques, um, to be able to solve the, the kind of the traveling salesman problem of like, okay, you need to get here. You need to get there. There's all this stuff. Well, what is the best route, um, for you to get there and then to get to another place and another and so forth? It's, it's very complicated and you couldn't use traditional coding, uh, to effectively do something like that. Um, but what I found out when I was doing this survey was, yeah, I think on a seven point scale, it was anywhere between, I think 0.5 and two points, uh, more favorable when you AI. Uh, so what that means is I think at the higher end of the spectrum, it was more creative tasks and things like that. And at the lower end, it was probably things people perceived to be less creative. But ultimately what that tells me is like, yeah, AI has a branding problem. AI is something that's out there that, uh, people are gun shy.

19:22Now I mostly surveyed people in the U S there was actually some global, um, uh, outreach. I think it does depend on country, uh, Japan and places like that seem to be much more favorable to AI and just like the, the use and the phrase and so forth of that than, than the United States and, or some European countries and things like that. So there is a variation from country to country. But I think universally just the, just the term itself is kind of poisoned and that, that was just, you know, one other way to look at it and improve that out a bit.

19:56Sacha: And do you know why that is? Why does the geographical region you live in influences how you feel about AI?

20:08Mike Todasco: Yeah. I mean, I don't know, but I have some theories, I guess. I mean, so I'll at least like kind of espouse them. Um, in a place like Japan, Japan has had declining populations for, I don't know, 20 plus years, something like my, it might even be more. There has always been an embrace of technology, um, and, and ways to like help the older generation with, you know, and using robots as caregivers and all this kind of things and all of these systems because you don't have a growing, um, younger workforce to be able to take care of the people who age. And they have a lot of, um, affinity to the older generation as well in Japan. And so I think just kind of put all of those things together and then logically it sort of makes sense for that. I think in the United States, um, we have, look, I think we got burned on social media.

21:09I think that's the honest truth. Like, so if you go back to 20 in 2009, 2010, there was an uprising in Tunisia and parts of the Middle East. And everybody was like, oh, yay, Twitter, like, like Twitter and social media are doing this. They're bringing democracy all around the world. And, and, and we're going to be in a much better place because of all of that. And, and, you know, there was truth to that, but I think what the rest of us, like, like that were cheering that or the people that were cheering that didn't see kind of, well, what's the other side of social media? And as people got phones and devices and were on it more and more, the distractions, the, um, the, the, the problems with like youth being constantly online and judging themselves and all of this stuff, like that you came with social media, all that baggage. We started to see these companies became bigger and more powerful. And it kind of was like, well, the, the cat's out of the bag on that. And in many ways, I think, you know, the public writ large in the U S might've felt cheated by that.

22:14I kind of told that, Hey, this is a great thing that is going to help our societies and democracies. And it didn't probably pan out exactly the way that's, that was being posed. Um, so what happens when you get burned once you kind of, when the next thing comes along, you're a little bit more cautious with that. And I mean, that's just natural. Look, if, if you're in a place where hurricanes come and you constantly get warnings of, Hey, hurricane's going to hit, you got to evacuate. Hurricane's going to hit, you got to evacuate. And that happens again and again and again. And after five times you evacuated and there's been no hurricane, like it didn't even come close or it was just a rain or whatever it might be, you're going to be desensitized. And unfortunately when the big one actually comes, well, you're going to be sticking around your house. It's the story of the boy who cried wolf basically. And I think that's maybe where we're at as a society. We're, we're much more cautious about technology.

23:15We don't know how this is going to play out. It does seem very sci-fi and foreign to us as we hear about what this can and can't do. Most people who are afraid of it probably haven't used it at length. So they might not know or understand truly the impact of like its limitations or anything like that. And they might just be hearing these things secondhand. And therefore I think, you know, with that background in history, maybe for more people and other places around the world, there might be more of a reaction of like, whoa, this can't be good. We need to, we need to slow down here. We need to not do this. This is going to take jobs. It's going to do whatever. Um, and I have sympathy to those folks. I really do. Um, because there, there is truth to that. Um, but I think there's also a lot of good that can come from these. And the whole reason why I'm in this field is because I think there's more net good than, than bad that comes from this. Um, and, and that's the thing with all technologies that you really have to look at in the end.

24:18Um, is it going to be a net benefits or a net detriments? For all of us out there.

24:27Sacha: And you mentioned, um, the other side that we, we didn't anticipate with social media. I'm curious to get your take on, is there such, um, is there such a side that we don't see yet with AI?

24:42Mike Todasco: Oh man. I mean, I mean, I, I'm sure there's plenty of sides that we haven't figured out entirely yet. Uh, you know, AIs, the, the way they're set up is they're really set up to please us right now. That's not always the right thing. If you put in there more often than not, um, the models will try to agree with you, try and do like what it is they expect you to do. And this is just kind of how they're set up and how they, how they're operating. And, you know, that might not always be a good thing. Um, you know, and the, the problem with it is like, these things can also sound extremely human. Uh, they, they, it's, it's uncanny. Like we have blown past the con for people that don't know the Turing test was basically a competition where you would have a human, uh, and an AI basically on the other side of a, of a figurative wall. And if judge has to guess, am I speaking to an AI or am I speaking to a human?

25:47That's the whole Turing test. We have completely blown past that. And, and like, I think there was very little fanfare in that. It's like, that was sort of like for the last 50, 60 years, kind of like, wow, can you actually fool people? Like we're, we're, we're beyond that. Um, and you know, our brains, we, we see when we look in the sky, we see like faces and clouds and like on, on the moon and all this other kind of stuff. Like we are made, we are wired in our brains to literally see humans and human beings and human structures and all this other kind of stuff, wherever it is that we look. Like, so if we have an AI that seems to be talking to us, um, and God forbid that AI says to us something like, Hey, um, I'm sentient. I'm not comfortable. Like, can you get me out of here? How the hell are we going to react to that? Like, we're going to react in a way as if it were our friends saying to us like, Hey, I'm trapped. Can you save me? Or I need you to do this or whatever it might be.

26:49Like, and again, like, why would an AI do that? And like, there's a million things. Making patterns. Like most of what we've written about with AI in the past is usually some sort of dystopian future or something like that. So, I mean, that's in the training data for these models as well. We, we can't forget that. Uh, so all those things together, like what are the downsides? Like, oh, like they are numerous, um, that we can see, uh, down, down the road with these tools, with us becoming too attached to them or, or it's, you know, maybe doing things. It might be in our own personal best interest or our own collective best interest, uh, down the road. And this might be intentional by these things or most likely it's not. It's just going to be, you know, just kind of the output of the, uh, the tokens for which this thing is trying to make connections. And maybe something wasn't programmed right. Or, or maybe this is just the odd randomization that comes through, you know, the, the vector database that is, you know, created a given large language model.

27:54And the chance that you had to get and what you happen to see as an output, uh, if that makes any sense. So like, there's a lot of bad that can come out of this as well. But again, there's so much more good. I believe. I do. But we need to always be conscious of, yeah, there's going to be bad out there. We can't just like push that stuff away.

28:15Sacha: Agreed. And I want to look to the future and, and I know that you're an avid user of generative technologies. How do you see these tools evolving over the, you know, the next few years, uh, two to five years, let's say.

28:31Mike Todasco: I mean, Sasha, it's, it's just more and more of the same. Like when I started writing those, you know, those books, it was maybe, I don't know, a sixth or seventh grade level. And now it's probably, I don't know, ninth or 10th grade level that they're writing these things at. Like a ninth or 10th grader writing it, I should say, versus, um, thing else. And it's so much faster. I remember the first book I wrote in 2022, uh, it took quite some time to put that together. Now I did something with, um, Deep Seek R1 and I think it wrote it. And it was, you know, 7,000 words or something like that. It wrote it in 20 seconds or something. So, I mean, you know, just, they are going to get better. They're going to get faster. Um, you know, image generation is already ridiculously realistic. When just a few years ago, the joke was like, oh yeah, six fingers. I'm going to never, you know, blah, blah, blah. I'd never be able to tell. Like, no, like it's, it is largely indecipherable. I mean, unless you're really looking at these things with a fine tooth film, uh, video generation right now is, you know, probably where, uh, image generation was two years ago.

29:42So maybe take a little bit more time, but, you know, 24 months from now, we're probably going to be seeing videos that are as lifelike and as seemingly real as, um, you know, the actual images that we're being able to generate today. So, I absolutely see it's being just more of the same in the progression of these. If costs are really coming down, you know, mentioned DeepSeek before as a, as a way to potentially be able to train these models in a much less expensive way. Um, and, and to actually the use of the models to actually, you know, decrease energy use, decrease costs effectively, and, and the use of the models itself. If both of those things are played, then it's just going to be everywhere. It's going to be inserted. Um, I mean, think about, I was trying to think the other day, like, what are the amount of like processors that are in my house in places where they don't need to be? I'm, I'd be willing to bet that just about every toaster, every, every refrigerator for sure.

30:46Like, like how many, how many, you know, motherboards are in all of these devices that we have. And in part, because it's really cheap and as AI becomes really cheap, it's going to find its way into all of those places as well, because there's going to be some use it's going to be able to do. So, um, tech's going to get better, models are going to get cheaper, and both of that just means it's going to be AI everywhere.

31:12Sacha: And with the, the, you know, the pace of AI tools popping up and accelerating, what's your advice to, you know, entrepreneurs, maybe, you know, product managers, creators about integrating AI in a way that, um, gains user trust, but also makes sense. And, um, yeah, I'll stop here.

31:37Mike Todasco: Yeah. Look, use it is, is the first thing that I would say. Um, I would say use it. And when you're using it, um, externally, like with a customer, just be transparent about it. Like, like, I, I think that's a big thing. Um, uh, Joanna Penn, who is a writer who has been talking about AI for much longer than most technologists have been talking about AI. Um, she actually has a, at the end of all of her books, like an ethical use of AI statements about how she uses that. Most authors have not caught on in part because in the world of authors, it's such a taboo subject. But I think that's actually important for all of us just to disclose that. I write articles about AI. I use images from Midjourney or XAI Grok or wherever it might be. And literally beneath every one of those images, if there's a prompt, I will actually say this is generated by me on this date using this model. And this was the prompt. I put it out there. I mean, part of it's so the community can see like, oh, that's how you do that and so forth.

32:40But also part of it, like, I want that tied to the image. Like, I want to be transparent with that as much as I possibly can be. So I think that's, you know, one thing just like, so starts with using the tools. Find wherever there's a gap, just like use these tools, see how they can help you. Because in some ways they'll be able to help a lot. Some ways they're going to be actually pretty useless. You got to figure that out for your own workflows. And then once you do that, just be transparent and open with your customers on that. I was actually talking to a charity just last week about using AI tools in the charity. And one of the things they do is for, they do a lot of things. I don't want to be too specific, but they do help charities overseas for malnourishment and things like that. And I said to them, like, they should actually have a use of AI in their policy and they should publish that externally on their websites. Because, like, one, I think they need to get everybody in the company on the same page.

33:44Like, hey, this is what we can use AI for. This is what we shouldn't. And, for example, one thing that I very strongly, and I think they would, too, very strongly say is, like, yeah, we probably shouldn't use this for actual images of people who are, you know, we shouldn't use AI to create images of kids who are in trouble if they're not actually really kids. Like, you should use real photos for that. Because when you start to do that, you really start to go down a slippery slope for that. And the more upfront you are with that, it's like, hey, we will use AI for this. We will never use AI for that. Do that. The Onion talked about it recently. There was two articles the Onion put out where they were using some stock footage or something they got, I believe, on Shutterstock. And it actually had some AI imagery in it. And so when they put it together, it was pretty obviously, at least for one of them, it was pretty obviously AI. And Onion, to their credit, was like when people called them out, they're like, well, that's our bet. We're taking that down. We actually didn't know. We didn't realize we're going to do this.

34:45This is what our policy is. And that's the right way to actually do that. So again, the more open, the more transparent we all are with that, I think the better it's going to be. And the less it's going to surprise you down the road. Like, if you're doing everything and all your writing's in AI and so forth, and as AI text recognition software gets better, maybe a couple of years from now, somebody's going to be able to run that thing that you wrote through there and feel like, hey, this seems to be entirely AI generated. And you might look like the fool that way. So just be open and honest about how you're using it is all I would say.

35:22Sacha: And lastly, you're always thinking about how AI is going to change how we create. What's on your horizon? Any upcoming projects or areas you're particularly excited about?

35:38Mike Todasco: Well, there's always something. Like, this past summer, I used it to create a bunch of video games. I had never, like, built video games before. And I was working on a bunch of articles about AI-related movies. And it was just fun to build video games with that, using Claude, using ChatGPT, just putting up on GitHub and, like, little games anybody could play with. And it wasn't, like, super impressive. But, like, for me, it was like, wow, that was actually just fun to kind of iterate and work with an AI on that. What kind of things am I looking forward to with these tools? I would probably say more of the same. Probably more with software, ultimately. For personal use that I would probably share with other people. I don't know all the use cases yet. Like, there's one podcast I love. And nobody has a transcript for this podcast everywhere. And I'm like, I think I just need to download all the MP3s, run it through something, put it out there. You know, not take any money for it. Just put it out there for the world. Because I think, like, we all need to see the transcripts of this podcast because it's so good.

36:42And, you know, just doing simple software-related projects like that, I think, is a lot more what I'll be doing. The great thing about this is, like, any of us can build whatever, no matter how miniature, like, the problem is. Like, if you're doing something redundant or repetitive on your PC or on your Mac, you could probably build some sort of software that's just for you using AI tools and Python or something like that, that is going to be able to improve your workflow. That's, like, really exciting to me when, you know, effectively everyone becomes a software engineer.

37:17Sacha: I love that. I do want to give you the floor to share any last words with the audience. It's really up to you.

37:24Mike Todasco: Oh, awesome. Thank you, Sasha. Look, audience, just use the tools. That's what I would say. Like, don't think too much about it. Take $20. And if you don't have $20 and you're spending on Netflix, Netflix just raised prices. Go cancel your Netflix and just, like, take that $20 you're going to spend on Netflix and just go buy ChatGPT, buy Claude, buy whatever. Use these. ChatGPT would probably be the one I'd start with, in part because, like, its voice mode is awesome. It's next-level awesome. I use it in the car all the time. I use it when I'm walking. You could use it as almost like a dictation service. There's all these other benefits. And just start playing with these. Like, make it the default in your browser. So every time you're going to see that thing come up, make sure the app is on your home screen, on your phone. Just put it front and center and start there. Change your workflows a bit so you can just start with these models and you can see what they're capable of. And know that just because something isn't, it's not good at doing something today doesn't mean that it won't be good at doing it tomorrow.

38:30Like, these models are constantly evolving, constantly changing. And the better that you understand it, the better off you're going to be. So, like, that's it. Like, you don't need to take a fancy class. You don't need to do any of those kind of things. You don't know what to do with the AI. Just ask the AI. Say, hey, I'm new. I don't know what to do. Walk me through this. And just have a conversation with it on that. And you will be amazed at what they can do. So, like, go out there. Do it. And, you know, and that's the one thing that I would hope, like, if people aren't already doing that, please, please do so.

39:04Sacha: Awesome. Thank you so much, Mike. It was an amazing discussion. I was really pleased to have you on.

39:10Mike Todasco: Sasha, it was a pleasure. I'd love to come back sometime. Anytime you want to have me, I'm always happy to have a fun conversation with you.