{"video_id":"fp_0pMBQAROL6","title":"What are your thoughts on AI?","channel":"FP Exclusives","show":"FP Exclusives","published_at":"2023-09-01T22:53:00.017Z","duration_s":6217,"segments":[{"start_s":0.0,"end_s":3.0,"text":"Hello, I am here with Emily Young.","speaker":null,"is_sponsor":0},{"start_s":3.0,"end_s":9.4,"text":"And what do you do for the company? I am a writer. I used to be a benchmarker, but that's lab's job now.","speaker":null,"is_sponsor":0},{"start_s":9.4,"end_s":14.44,"text":"Excellent, excellent. So I'm just asking everyone in the company here what their thoughts on AI is, like the state of it.","speaker":null,"is_sponsor":0},{"start_s":14.44,"end_s":18.36,"text":"So what are your thoughts? We don't have it.","speaker":null,"is_sponsor":0},{"start_s":18.36,"end_s":23.04,"text":"No, it's not. What we refer to as AI right now,","speaker":null,"is_sponsor":0},{"start_s":23.04,"end_s":27.24,"text":"I think it would be better to describe it as like,","speaker":null,"is_sponsor":0},{"start_s":27.24,"end_s":31.28,"text":"have you ever played Mass Effect?","speaker":null,"is_sponsor":0},{"start_s":31.28,"end_s":35.12,"text":"Two, but a long time ago. OK, so in the lore of Mass Effect,","speaker":null,"is_sponsor":0},{"start_s":35.12,"end_s":39.64,"text":"AIs are like, you don't do that.","speaker":null,"is_sponsor":0},{"start_s":39.64,"end_s":43.96,"text":"So instead, what they have is what's called a VI, a virtual intelligence.","speaker":null,"is_sponsor":0},{"start_s":43.96,"end_s":49.68,"text":"And that is a lot like what we consider to be like chatGPT and all that kind of stuff right now.","speaker":null,"is_sponsor":0},{"start_s":49.68,"end_s":52.88,"text":"So it's like, it's able to answer questions.","speaker":null,"is_sponsor":0},{"start_s":52.88,"end_s":55.88,"text":"It's able to carry a conversation.","speaker":null,"is_sponsor":0},{"start_s":55.92,"end_s":59.56,"text":"It's able to do a lot of things, but it's not able to think.","speaker":null,"is_sponsor":0},{"start_s":59.56,"end_s":67.56,"text":"It doesn't have actual thoughts or feelings. So what it's doing is, in terms of like chatGPT or whatever,","speaker":null,"is_sponsor":0},{"start_s":67.56,"end_s":71.08,"text":"it's basically, it's learning patterns.","speaker":null,"is_sponsor":0},{"start_s":71.08,"end_s":76.88,"text":"So it's just learning patterns and spinning out what it thinks is the next thing in the pattern, more or less.","speaker":null,"is_sponsor":0},{"start_s":76.88,"end_s":82.92,"text":"At least that's my overly simplistic explanation of it.","speaker":null,"is_sponsor":0},{"start_s":85.92,"end_s":89.72,"text":"So I don't think we have AI in the traditional sense right now.","speaker":null,"is_sponsor":0},{"start_s":89.72,"end_s":95.8,"text":"At least that's, I'm talking about generative AI that people think is what AI is.","speaker":null,"is_sponsor":0},{"start_s":95.8,"end_s":101.4,"text":"But it can be pretty convincing. So then it sounds like you're kind of not liking AI.","speaker":null,"is_sponsor":0},{"start_s":101.4,"end_s":105.36,"text":"Is that right to assume or?","speaker":null,"is_sponsor":0},{"start_s":105.36,"end_s":109.48,"text":"AI is difficult. There are things that AI is really good for.","speaker":null,"is_sponsor":0},{"start_s":109.48,"end_s":113.2,"text":"I mean, it's basically like, say for example, AI image","speaker":null,"is_sponsor":0},{"start_s":113.2,"end_s":117.2,"text":"scaling or whatever. You train it to do whatever it needs to do, and it does it.","speaker":null,"is_sponsor":0},{"start_s":117.2,"end_s":120.92,"text":"Basically by failing as many times as is necessary","speaker":null,"is_sponsor":0},{"start_s":120.92,"end_s":124.44,"text":"to fail the correct way.","speaker":null,"is_sponsor":0},{"start_s":124.44,"end_s":129.56,"text":"And that's actually really cool, because it does things in a way that is very difficult to do","speaker":null,"is_sponsor":0},{"start_s":129.56,"end_s":133.08,"text":"in traditional software.","speaker":null,"is_sponsor":0},{"start_s":133.08,"end_s":137.16,"text":"When you get into marketing AI as, well, I mean,","speaker":null,"is_sponsor":0},{"start_s":137.16,"end_s":142.0,"text":"I keep saying AI. It's more like machine learning. So when you get into marketing it as like AI,","speaker":null,"is_sponsor":0},{"start_s":142.04,"end_s":146.08,"text":"and you say, oh, well, chatGPT is going to take over the world, well, that's overselling it.","speaker":null,"is_sponsor":0},{"start_s":146.08,"end_s":150.28,"text":"And I think it's by design that they're doing that,","speaker":null,"is_sponsor":0},{"start_s":150.28,"end_s":154.2,"text":"because they want people to invest. They want people to think that we're close enough,","speaker":null,"is_sponsor":0},{"start_s":154.2,"end_s":157.72,"text":"and that these technology companies that are investing in it","speaker":null,"is_sponsor":0},{"start_s":157.72,"end_s":160.44,"text":"are more powerful than they really are.","speaker":null,"is_sponsor":0},{"start_s":163.32,"end_s":169.0,"text":"But that being said, AI is a tool like any other tool. It can be used for good or evil.","speaker":null,"is_sponsor":0},{"start_s":169.0,"end_s":173.72,"text":"There are ways to generate images, for example, that are not copyright infringing.","speaker":null,"is_sponsor":0},{"start_s":173.72,"end_s":177.92,"text":"I think Adobe's is potentially non-infringing,","speaker":null,"is_sponsor":0},{"start_s":177.92,"end_s":183.12,"text":"but I know that stable diffusion and what have you are,","speaker":null,"is_sponsor":0},{"start_s":183.12,"end_s":186.28,"text":"as far as copyright goes.","speaker":null,"is_sponsor":0},{"start_s":186.28,"end_s":191.28,"text":"Not tested in courts, obviously, but if you take somebody's art and you put it into the generator","speaker":null,"is_sponsor":0},{"start_s":191.28,"end_s":196.04,"text":"and it just blends it up with everything else, then what is that?","speaker":null,"is_sponsor":0},{"start_s":196.04,"end_s":201.76,"text":"That's the question that we have to answer. And so far, we don't have one.","speaker":null,"is_sponsor":0},{"start_s":201.76,"end_s":206.96,"text":"So yeah, it's not a good or bad thing. It's just that the way we use it is really","speaker":null,"is_sponsor":0},{"start_s":206.96,"end_s":210.32,"text":"what determines whether it's good or bad.","speaker":null,"is_sponsor":0},{"start_s":210.32,"end_s":214.48,"text":"OK, OK. So then as I creative, you do write.","speaker":null,"is_sponsor":0},{"start_s":214.48,"end_s":219.56,"text":"And I think one of the things people talk about when AI, what topic of AI is, is that, oh,","speaker":null,"is_sponsor":0},{"start_s":219.56,"end_s":224.56,"text":"it's going to take over writer's jobs. Do you believe that? Do you think that's a thing in the distant future?","speaker":null,"is_sponsor":0},{"start_s":224.56,"end_s":228.84,"text":"Or do you think it's inevitable in the next five years like that, or?","speaker":null,"is_sponsor":0},{"start_s":228.84,"end_s":235.04,"text":"The problem with that is that if no new writing is made,","speaker":null,"is_sponsor":0},{"start_s":235.04,"end_s":238.16,"text":"like if, say, for example, writing is completely no longer","speaker":null,"is_sponsor":0},{"start_s":238.16,"end_s":243.32,"text":"a thing because AI is taken over, what you then end up with","speaker":null,"is_sponsor":0},{"start_s":243.32,"end_s":247.96,"text":"is, what do they call it? Like noise amplification or generational noise.","speaker":null,"is_sponsor":0},{"start_s":247.96,"end_s":250.96,"text":"That basically, as you iterate on things that are already","speaker":null,"is_sponsor":0},{"start_s":250.96,"end_s":255.48,"text":"iterated on, you kind of lose bits and pieces of it.","speaker":null,"is_sponsor":0},{"start_s":255.52,"end_s":261.48,"text":"So it'll be difficult to continue indefinitely.","speaker":null,"is_sponsor":0},{"start_s":261.48,"end_s":266.0,"text":"You might get away with it a little bit for a little while. And people are already getting away","speaker":null,"is_sponsor":0},{"start_s":266.0,"end_s":269.56,"text":"with writing paragraphs or whatever with chat","speaker":null,"is_sponsor":0},{"start_s":269.56,"end_s":276.56,"text":"GPT or what have you. But it still requires an editing pass.","speaker":null,"is_sponsor":0},{"start_s":276.56,"end_s":282.24,"text":"It still requires people who know what they're doing because I don't know if you or people who are watching know.","speaker":null,"is_sponsor":0},{"start_s":282.28,"end_s":287.2,"text":"But AI, especially like chat GPT, has a very short memory.","speaker":null,"is_sponsor":0},{"start_s":287.2,"end_s":292.52,"text":"So if you're trying to tell a story, like in a movie or something, keeping that thing on the rails","speaker":null,"is_sponsor":0},{"start_s":292.52,"end_s":299.04,"text":"is going to be a really tough operation.","speaker":null,"is_sponsor":0},{"start_s":299.04,"end_s":304.56,"text":"So you basically have to keep feeding it like the background information because, what do they call it?","speaker":null,"is_sponsor":0},{"start_s":304.56,"end_s":310.92,"text":"I forget what they call it. But basically, you only have a certain number of basically","speaker":null,"is_sponsor":0},{"start_s":310.92,"end_s":315.24,"text":"things that it can remember indefinitely.","speaker":null,"is_sponsor":0},{"start_s":315.24,"end_s":319.48,"text":"And once you run out, it starts dropping them","speaker":null,"is_sponsor":0},{"start_s":319.48,"end_s":323.0,"text":"and adds more. And eventually, you have like, oh, well,","speaker":null,"is_sponsor":0},{"start_s":323.0,"end_s":327.52,"text":"you started out in New Zealand. And now you're in, I don't know, Turkey","speaker":null,"is_sponsor":0},{"start_s":327.52,"end_s":335.28,"text":"because it just chose another location because it forgot what location it was, for example.","speaker":null,"is_sponsor":0},{"start_s":335.28,"end_s":338.6,"text":"I've tried using chat GPT to write code.","speaker":null,"is_sponsor":0},{"start_s":338.6,"end_s":341.68,"text":"And it's really, really handy for small bits of code.","speaker":null,"is_sponsor":0},{"start_s":341.68,"end_s":345.32,"text":"But if you try to get it to do something substantial","speaker":null,"is_sponsor":0},{"start_s":345.32,"end_s":348.68,"text":"with that code, no.","speaker":null,"is_sponsor":0},{"start_s":348.68,"end_s":351.84,"text":"Yeah, that's kind of an assumption. Like I kind of give you a lot of people prior.","speaker":null,"is_sponsor":0},{"start_s":351.84,"end_s":355.4,"text":"And it's a good starting point, I guess.","speaker":null,"is_sponsor":0},{"start_s":355.4,"end_s":359.6,"text":"But then you need to have an actual person to look over it","speaker":null,"is_sponsor":0},{"start_s":359.6,"end_s":362.6,"text":"and expand on it further after that. Yeah.","speaker":null,"is_sponsor":0},{"start_s":362.6,"end_s":367.04,"text":"And obviously, it's going to get better. And it's going to be able to handle more and more things.","speaker":null,"is_sponsor":0},{"start_s":367.04,"end_s":370.72,"text":"And I mean, eventually, it might get to the point where it could write a screenplay or an operating","speaker":null,"is_sponsor":0},{"start_s":370.72,"end_s":377.08,"text":"system or something. The problem then becomes, well, how","speaker":null,"is_sponsor":0},{"start_s":377.08,"end_s":384.16,"text":"do we determine the copyright of that? How do we determine whether it's all actually good?","speaker":null,"is_sponsor":0},{"start_s":384.16,"end_s":396.68,"text":"Sometimes the code that it spits out isn't super conventional.","speaker":null,"is_sponsor":0},{"start_s":396.72,"end_s":403.28,"text":"So if it spits out something that is completely and totally unintelligible, and it turns out that there's a bug later on,","speaker":null,"is_sponsor":0},{"start_s":403.28,"end_s":407.92,"text":"like fixing that bug becomes a little difficult.","speaker":null,"is_sponsor":0},{"start_s":407.92,"end_s":414.64,"text":"You could use AI to fix that bug then. But then you have compounding factors","speaker":null,"is_sponsor":0},{"start_s":414.64,"end_s":418.16,"text":"that you're not quite sure how the system even works.","speaker":null,"is_sponsor":0},{"start_s":418.16,"end_s":422.64,"text":"So I think that for the foreseeable future,","speaker":null,"is_sponsor":0},{"start_s":422.64,"end_s":425.64,"text":"there's at least going to need to be some kind of review","speaker":null,"is_sponsor":0},{"start_s":425.68,"end_s":430.12,"text":"process. So you might build a function in AI.","speaker":null,"is_sponsor":0},{"start_s":430.12,"end_s":433.28,"text":"And then you can put that in. Then you build another function. You put that in.","speaker":null,"is_sponsor":0},{"start_s":433.28,"end_s":438.44,"text":"And basically, you stitch it together yourself, right? That's probably the best way to do it.","speaker":null,"is_sponsor":0},{"start_s":438.44,"end_s":441.44,"text":"Yeah. Yeah.","speaker":null,"is_sponsor":0},{"start_s":441.44,"end_s":446.88,"text":"Sorry. OK, so then as we are a tech company and a creative company,","speaker":null,"is_sponsor":0},{"start_s":446.88,"end_s":451.44,"text":"do you think that we should be using AI at any level? Or do you think we should be like,","speaker":null,"is_sponsor":0},{"start_s":451.44,"end_s":456.68,"text":"we don't want to touch that at all? Depends on the AI. And depends on what we're using it for.","speaker":null,"is_sponsor":0},{"start_s":456.68,"end_s":463.92,"text":"If we're talking chat GPT, I don't know that it's not copyright infringing.","speaker":null,"is_sponsor":0},{"start_s":463.92,"end_s":469.2,"text":"So that's one thing. The fact that we don't know what the actual data sets","speaker":null,"is_sponsor":0},{"start_s":469.2,"end_s":475.52,"text":"that they train them on are is a bit of a big, giant question","speaker":null,"is_sponsor":0},{"start_s":475.52,"end_s":478.68,"text":"mark in terms of the ethics and the, I guess,","speaker":null,"is_sponsor":0},{"start_s":478.68,"end_s":486.12,"text":"legality, ultimately, of using AI for generative purposes.","speaker":null,"is_sponsor":0},{"start_s":486.12,"end_s":493.24,"text":"On the other hand, say, for example, I'm like 95% of the way through a script.","speaker":null,"is_sponsor":0},{"start_s":493.24,"end_s":496.44,"text":"And you're like, you don't know how to close it out.","speaker":null,"is_sponsor":0},{"start_s":496.44,"end_s":500.2,"text":"You could sit there and you could scratch your head for an hour or two.","speaker":null,"is_sponsor":0},{"start_s":500.2,"end_s":505.88,"text":"Or you could ping some concepts off of chat GPT or something","speaker":null,"is_sponsor":0},{"start_s":505.88,"end_s":511.48,"text":"and see what it looks like. And I think that is the best way to go for it.","speaker":null,"is_sponsor":0},{"start_s":511.48,"end_s":516.04,"text":"So say if you were a visual artist, for example,","speaker":null,"is_sponsor":0},{"start_s":516.04,"end_s":521.8,"text":"what I would say is a relatively benign use of stable diffusion","speaker":null,"is_sponsor":0},{"start_s":521.8,"end_s":527.04,"text":"or what have you would be is laying out a scene","speaker":null,"is_sponsor":0},{"start_s":527.04,"end_s":530.32,"text":"or something like that. So yeah, I want to have this here and this here and this here","speaker":null,"is_sponsor":0},{"start_s":530.32,"end_s":536.04,"text":"and this is what the perspective should look like. And then when you have that, then you can paint over it.","speaker":null,"is_sponsor":0},{"start_s":536.04,"end_s":542.04,"text":"So I feel like that's the kind of thing that ultimately we should be using it for.","speaker":null,"is_sponsor":0},{"start_s":542.04,"end_s":545.96,"text":"I know that many people will be using it for and already are","speaker":null,"is_sponsor":0},{"start_s":545.96,"end_s":550.64,"text":"using it for complete works, which is a little less","speaker":null,"is_sponsor":0},{"start_s":550.64,"end_s":555.88,"text":"defensible. Yeah, I think that Disney's sequence invasion","speaker":null,"is_sponsor":0},{"start_s":555.88,"end_s":560.4,"text":"was a big example of like, because from what I am aware,","speaker":null,"is_sponsor":0},{"start_s":560.4,"end_s":565.72,"text":"they just put it in the prompt and that's it, right? They didn't change anything, right?","speaker":null,"is_sponsor":0},{"start_s":566.16,"end_s":570.6,"text":"I'm not sure. Yeah, OK. Yeah, I haven't seen that. Yeah, I just heard about it and I was like, oh.","speaker":null,"is_sponsor":0},{"start_s":570.6,"end_s":574.44,"text":"But then I wasn't like that invested during that time, but I was like the last two weeks.","speaker":null,"is_sponsor":0},{"start_s":574.44,"end_s":579.88,"text":"I'm like, I'm getting more into it, but. Yeah, I mean, the few things that I've seen, like, oh, well,","speaker":null,"is_sponsor":0},{"start_s":579.88,"end_s":584.44,"text":"AI is going to completely just take over the movie industry. Every frame was generated of this thing.","speaker":null,"is_sponsor":0},{"start_s":584.44,"end_s":589.28,"text":"I don't remember what it was, but it's like, it looks not good yet.","speaker":null,"is_sponsor":0},{"start_s":589.28,"end_s":595.64,"text":"And I guess you can see that it's there, sort of. But it doesn't really have, I guess,","speaker":null,"is_sponsor":0},{"start_s":595.64,"end_s":600.08,"text":"the memory to create a plot, and it still","speaker":null,"is_sponsor":0},{"start_s":600.08,"end_s":603.32,"text":"can't do fingers quite perfectly.","speaker":null,"is_sponsor":0},{"start_s":603.32,"end_s":612.12,"text":"So you've got like after images of stuff, you've got like six fingers or 12, stuff like that,","speaker":null,"is_sponsor":0},{"start_s":612.12,"end_s":616.72,"text":"or like teeth on teeth kind of thing, where it thinks stuff should go because that's","speaker":null,"is_sponsor":0},{"start_s":616.72,"end_s":622.12,"text":"where the pattern is, right, the pattern that it was fed. So it's like, OK, so teeth go here.","speaker":null,"is_sponsor":0},{"start_s":622.12,"end_s":629.2,"text":"And oh, yeah, that looks like teeth might go there, too. And right, so that's kind of how that works.","speaker":null,"is_sponsor":0},{"start_s":629.2,"end_s":633.04,"text":"So we're just going to wrap it up here. Is there anything that we didn't talk about","speaker":null,"is_sponsor":0},{"start_s":633.04,"end_s":636.64,"text":"that you wanted to bring it up, or just want to say anything to the float planers?","speaker":null,"is_sponsor":0},{"start_s":636.64,"end_s":640.88,"text":"I'm trying to coin that term now. People who subscribe, float planers, yeah.","speaker":null,"is_sponsor":0},{"start_s":640.88,"end_s":644.24,"text":"You think it's a good name? That's what we've been calling them internally for a while. Really?","speaker":null,"is_sponsor":0},{"start_s":644.24,"end_s":647.64,"text":"Yeah, float planers. All right, float planers, you heard it first.","speaker":null,"is_sponsor":0},{"start_s":648.08,"end_s":652.68,"text":"Is there anything you want to say about the topic, or anything you want to mention in general, or?","speaker":null,"is_sponsor":0},{"start_s":652.68,"end_s":656.88,"text":"Just reiterate that everything, all tools,","speaker":null,"is_sponsor":0},{"start_s":656.88,"end_s":659.88,"text":"are inherently not good or evil, right?","speaker":null,"is_sponsor":0},{"start_s":659.88,"end_s":663.2,"text":"It is what you use it for that is the most important.","speaker":null,"is_sponsor":0},{"start_s":663.2,"end_s":666.88,"text":"So AI image upscaling is amazing, for example.","speaker":null,"is_sponsor":0},{"start_s":666.88,"end_s":670.64,"text":"It's just one of those things that's just so universally","speaker":null,"is_sponsor":0},{"start_s":670.64,"end_s":674.8,"text":"useful that as it gets better, it","speaker":null,"is_sponsor":0},{"start_s":674.8,"end_s":680.04,"text":"becomes less and less likely that we'll ever not use it.","speaker":null,"is_sponsor":0},{"start_s":680.04,"end_s":683.96,"text":"Whereas those generative techniques, I don't know.","speaker":null,"is_sponsor":0},{"start_s":683.96,"end_s":688.52,"text":"It's one of those things where we might actually end up saying, OK, so we need to take a step back","speaker":null,"is_sponsor":0},{"start_s":688.52,"end_s":693.4,"text":"and ask ourselves some tough questions about how we want to do this.","speaker":null,"is_sponsor":0},{"start_s":693.4,"end_s":698.48,"text":"As far as AI, the concept of actual artificial intelligence","speaker":null,"is_sponsor":0},{"start_s":698.48,"end_s":704.8,"text":"goes, where are ways out from that, from what I understand?","speaker":null,"is_sponsor":0},{"start_s":704.8,"end_s":711.08,"text":"But I mean, we can simulate it pretty well with what we have now. And soon, it might not even matter.","speaker":null,"is_sponsor":0},{"start_s":711.08,"end_s":715.2,"text":"If you simulate it close enough, then what's the difference, right?","speaker":null,"is_sponsor":0},{"start_s":715.2,"end_s":720.16,"text":"So I guess the ultimate takeaway is be nice to your computer.","speaker":null,"is_sponsor":0},{"start_s":720.16,"end_s":724.88,"text":"Shut it down properly. You heard it first, don't unplug it randomly.","speaker":null,"is_sponsor":0},{"start_s":724.88,"end_s":728.12,"text":"All right, thank you very much, Emily, for your time. Thank you.","speaker":null,"is_sponsor":0}],"full_text":"Hello, I am here with Emily Young. And what do you do for the company? I am a writer. I used to be a benchmarker, but that's lab's job now. Excellent, excellent. So I'm just asking everyone in the company here what their thoughts on AI is, like the state of it. So what are your thoughts? We don't have it. No, it's not. What we refer to as AI right now, I think it would be better to describe it as like, have you ever played Mass Effect? Two, but a long time ago. OK, so in the lore of Mass Effect, AIs are like, you don't do that. So instead, what they have is what's called a VI, a virtual intelligence. And that is a lot like what we consider to be like chatGPT and all that kind of stuff right now. So it's like, it's able to answer questions. It's able to carry a conversation. It's able to do a lot of things, but it's not able to think. It doesn't have actual thoughts or feelings. So what it's doing is, in terms of like chatGPT or whatever, it's basically, it's learning patterns. So it's just learning patterns and spinning out what it thinks is the next thing in the pattern, more or less. At least that's my overly simplistic explanation of it. So I don't think we have AI in the traditional sense right now. At least that's, I'm talking about generative AI that people think is what AI is. But it can be pretty convincing. So then it sounds like you're kind of not liking AI. Is that right to assume or? AI is difficult. There are things that AI is really good for. I mean, it's basically like, say for example, AI image scaling or whatever. You train it to do whatever it needs to do, and it does it. Basically by failing as many times as is necessary to fail the correct way. And that's actually really cool, because it does things in a way that is very difficult to do in traditional software. When you get into marketing AI as, well, I mean, I keep saying AI. It's more like machine learning. So when you get into marketing it as like AI, and you say, oh, well, chatGPT is going to take over the world, well, that's overselling it. And I think it's by design that they're doing that, because they want people to invest. They want people to think that we're close enough, and that these technology companies that are investing in it are more powerful than they really are. But that being said, AI is a tool like any other tool. It can be used for good or evil. There are ways to generate images, for example, that are not copyright infringing. I think Adobe's is potentially non-infringing, but I know that stable diffusion and what have you are, as far as copyright goes. Not tested in courts, obviously, but if you take somebody's art and you put it into the generator and it just blends it up with everything else, then what is that? That's the question that we have to answer. And so far, we don't have one. So yeah, it's not a good or bad thing. It's just that the way we use it is really what determines whether it's good or bad. OK, OK. So then as I creative, you do write. And I think one of the things people talk about when AI, what topic of AI is, is that, oh, it's going to take over writer's jobs. Do you believe that? Do you think that's a thing in the distant future? Or do you think it's inevitable in the next five years like that, or? The problem with that is that if no new writing is made, like if, say, for example, writing is completely no longer a thing because AI is taken over, what you then end up with is, what do they call it? Like noise amplification or generational noise. That basically, as you iterate on things that are already iterated on, you kind of lose bits and pieces of it. So it'll be difficult to continue indefinitely. You might get away with it a little bit for a little while. And people are already getting away with writing paragraphs or whatever with chat GPT or what have you. But it still requires an editing pass. It still requires people who know what they're doing because I don't know if you or people who are watching know. But AI, especially like chat GPT, has a very short memory. So if you're trying to tell a story, like in a movie or something, keeping that thing on the rails is going to be a really tough operation. So you basically have to keep feeding it like the background information because, what do they call it? I forget what they call it. But basically, you only have a certain number of basically things that it can remember indefinitely. And once you run out, it starts dropping them and adds more. And eventually, you have like, oh, well, you started out in New Zealand. And now you're in, I don't know, Turkey because it just chose another location because it forgot what location it was, for example. I've tried using chat GPT to write code. And it's really, really handy for small bits of code. But if you try to get it to do something substantial with that code, no. Yeah, that's kind of an assumption. Like I kind of give you a lot of people prior. And it's a good starting point, I guess. But then you need to have an actual person to look over it and expand on it further after that. Yeah. And obviously, it's going to get better. And it's going to be able to handle more and more things. And I mean, eventually, it might get to the point where it could write a screenplay or an operating system or something. The problem then becomes, well, how do we determine the copyright of that? How do we determine whether it's all actually good? Sometimes the code that it spits out isn't super conventional. So if it spits out something that is completely and totally unintelligible, and it turns out that there's a bug later on, like fixing that bug becomes a little difficult. You could use AI to fix that bug then. But then you have compounding factors that you're not quite sure how the system even works. So I think that for the foreseeable future, there's at least going to need to be some kind of review process. So you might build a function in AI. And then you can put that in. Then you build another function. You put that in. And basically, you stitch it together yourself, right? That's probably the best way to do it. Yeah. Yeah. Sorry. OK, so then as we are a tech company and a creative company, do you think that we should be using AI at any level? Or do you think we should be like, we don't want to touch that at all? Depends on the AI. And depends on what we're using it for. If we're talking chat GPT, I don't know that it's not copyright infringing. So that's one thing. The fact that we don't know what the actual data sets that they train them on are is a bit of a big, giant question mark in terms of the ethics and the, I guess, legality, ultimately, of using AI for generative purposes. On the other hand, say, for example, I'm like 95% of the way through a script. And you're like, you don't know how to close it out. You could sit there and you could scratch your head for an hour or two. Or you could ping some concepts off of chat GPT or something and see what it looks like. And I think that is the best way to go for it. So say if you were a visual artist, for example, what I would say is a relatively benign use of stable diffusion or what have you would be is laying out a scene or something like that. So yeah, I want to have this here and this here and this here and this is what the perspective should look like. And then when you have that, then you can paint over it. So I feel like that's the kind of thing that ultimately we should be using it for. I know that many people will be using it for and already are using it for complete works, which is a little less defensible. Yeah, I think that Disney's sequence invasion was a big example of like, because from what I am aware, they just put it in the prompt and that's it, right? They didn't change anything, right? I'm not sure. Yeah, OK. Yeah, I haven't seen that. Yeah, I just heard about it and I was like, oh. But then I wasn't like that invested during that time, but I was like the last two weeks. I'm like, I'm getting more into it, but. Yeah, I mean, the few things that I've seen, like, oh, well, AI is going to completely just take over the movie industry. Every frame was generated of this thing. I don't remember what it was, but it's like, it looks not good yet. And I guess you can see that it's there, sort of. But it doesn't really have, I guess, the memory to create a plot, and it still can't do fingers quite perfectly. So you've got like after images of stuff, you've got like six fingers or 12, stuff like that, or like teeth on teeth kind of thing, where it thinks stuff should go because that's where the pattern is, right, the pattern that it was fed. So it's like, OK, so teeth go here. And oh, yeah, that looks like teeth might go there, too. And right, so that's kind of how that works. So we're just going to wrap it up here. Is there anything that we didn't talk about that you wanted to bring it up, or just want to say anything to the float planers? I'm trying to coin that term now. People who subscribe, float planers, yeah. You think it's a good name? That's what we've been calling them internally for a while. Really? Yeah, float planers. All right, float planers, you heard it first. Is there anything you want to say about the topic, or anything you want to mention in general, or? Just reiterate that everything, all tools, are inherently not good or evil, right? It is what you use it for that is the most important. So AI image upscaling is amazing, for example. It's just one of those things that's just so universally useful that as it gets better, it becomes less and less likely that we'll ever not use it. Whereas those generative techniques, I don't know. It's one of those things where we might actually end up saying, OK, so we need to take a step back and ask ourselves some tough questions about how we want to do this. As far as AI, the concept of actual artificial intelligence goes, where are ways out from that, from what I understand? But I mean, we can simulate it pretty well with what we have now. And soon, it might not even matter. If you simulate it close enough, then what's the difference, right? So I guess the ultimate takeaway is be nice to your computer. Shut it down properly. You heard it first, don't unplug it randomly. All right, thank you very much, Emily, for your time. Thank you."}