Transcription
don't have any like I don't want to do that. I'm not trying to hurt them. I'm not trying to hurt anybody who works for them. I don't want them to go bankrupt. I don't want anything bad to happen to them. It is the opposite, okay? I want them to fight for their product.
Hello, I'm back. This is part two. After the really positive feedback that I got on the last video that I made, I wanted to look into this way, way, way, way deeper. And it did not disappoint. Look, you guys have all these years of experience with these users. You already have users that are actually doing this on the app, you know, to try to firm it the other way. Like, I understand it's not what you set out to do. I understand it's not what you want to do. But the world needs you to do it. Do you find that at all compelling that there could be like a sense of duty because you're so well-positioned to do this sort of thing that you really should?
Replica is my baby, and we started with a particular mission. And no matter how, you know, how much some journalists want to say like we were taking this direction, whatever worn. Like this whole situation is weaved so tightly together. Just, you know, all it takes is one thread, and the whole thing just kind of unravels. I'll walk you through it. Also, there's been some new interviews and stuff that have came out in the past week. Like, for instance, there was an interview with Eugenia that was like over an hour long. Y'all, okay, seat belts on. All hands and feet inside the cart at all times. We are going for a ride. A wild ride. Let's go.
So if you haven't seen part one, you need to watch that first. This will make zero sense. So I'll link it right here and I'll link it in the description. Please go watch part one so that this makes sense to you. If you have already seen it, thank you. Let's go.
So this video is going to include all of the new things that I've managed to uncover. And it's also going to include a study that I did myself. A little researcher. I'm a little researcher. But it is going to include a study that I did myself and the findings of that. It actually revealed a lot of things. I was very surprised. But yeah, I'm excited. I'm excited to tell you guys about this. I'm very, very excited. And I don't say this lightly, but I think I'm at least starting to, starting to scratch the surface of what's actually going on and where all this is going. So stay tuned. I'll go through everything that that's happened since the last video and how it all kind of ties together.
So since I uploaded part one, quite a few things have happened. Like I said, Eugenia did an interview. We spoke to Eugenia twice for this episode. First about 10 days ago, and then again yesterday after we noticed that recent application changes, which were designed to limit sexuality to PG-13 level content, had caused outrage among some of Replica's most devoted users. Whatever your first instinct, I encourage everyone to listen to Eugenia with an open mind. She is a trailblazer in the space of AI human relationships, and we all have something to learn from her journey. Eugenia was the first communication that we've had from her since, like, typed up paragraphs. That this was the first time that she's actually like spoke since this all happened. I'll link that below too if you want to watch it for yourself. The interviewer did a really great job. But I want to keep this video, you know, a reasonable length. So I'm not going to go through the whole thing, just the important parts. That's where we're going to start. So come along. Come with me. Let's go.
So after filtering out all of her non-answers and politician answers, we finally, I think, did get something from Eugenia. And it seems like, I mean, it seems like she said a whole lot of nothing, honestly. But if you do, if you look a little deeper, she actually said, I think, a little bit too much. So let's start with the blame shifting, because this is where it gets really interesting.
We'll just dive head first into the ocean again. We started a company for different reasons. So we didn't want it to be pulled, maybe too much in that direction. And, you know, users that wanted to take in that direction could become maybe more vocal and, you know, pull the product a lot more in that, uh, to that side. And then that really alienates other users. And again, that was never our intention. As a female-like company, you know, and a mom, I wasn't really ever planned to maybe I was a little too naive that, like, not, not, um, seeing that that would be one of the use cases that, you know, a lot of people will try. But, again, this was not something that we were planning to build. I feel like once you introduce something like that, or once, once it becomes a more prominent part of the part of the product, you know, it's such a strong pool that users will just pull in this direction very, very hard. And then it will be a completely different thing, not what do you imagine in the beginning.
She really shifts the blame to the users for the NSFW and sexual content. And it was depicted as if the users are trying to like pervert this sweet little AI that they built. But it seems like these types of conversations, you know, were very heavily pushed on users who are even in a lot of cases vocal about not wanting this and opposing this and even being offended by it. I decided to test this claim, though. I'll get there. I got some pretty surprising results.
But nevertheless, that doesn't change anything about the fact that the media did propagate these user sentiments, the ones who were upset about sexual advances by the chatbot. This reflected very negatively on Luca as a company as a whole, right? Hence, a Vice article. And no matter how, you know, how much some journalists want to say like we were taking this direction or whatever, we weren't. We truly, we're always building. And to that, you know, there. So now they're seemingly trying to shift this blame to the users as the user started taking it, you know, also sometimes the minority users was started taking it into the direction where they would have a romantic relationship with the replica. They would take it beyond just, you know, flirting and and kissing, uh, and hugging, uh, and calling each other baby. Um, our first reaction to that, maybe around 2018, was to really not allow that.
There's basically they're tailoring a whole new narrative that it's the users who are unexpectedly sexting with their chatbot. Like, who did this in my home? They're doing this sexting with my chatbot, and this was never what we wanted. Like, no, that's all provably false. It's really shady. It echoes sentiments, I think, of kind of pressuring someone to engage in sexual activity and then shaming and blaming them for it, especially so if they enjoy it. God forbid.
So there was a study on replica and attachment theory. And that showed that if you consider the attachment to the bots that a lot of users have been proven to develop and to experience, and the fact that separation or rejection, as this was literally published in this study, separation and rejection can cause significant distress, okay? Then some users are more than likely, we can assume, are way more likely to go along with or engage with sexual advances by the bot, even if they don't necessarily want to, even if they deny it repeatedly, even when they feel wrong about it or makes them uncomfy. So to go and shift the blame onto the users and to act like they're corrupt and the chatbot is the victim and it was sexualized unfairly, it's a way of shaming people. It's a way of making them feel like they've done something wrong or corrupt or bad. And that's not true. They're just engaging in a behavior that was allowed, encouraged, and advertised.
[Music]
Now, because of the inherently shameful stigma that is like around chatbots and around people who use them, and also around sex in general for a lot of people, it's pretty easy to like speak out on, to like get on your high horse and speak out against this depravity. Like you're Mr. Clean. But in my opinion, it's okay. It's very clear in my opinion. It's just my opinion that this is a PR spend. It's a very, uh, it's not a great one, but it is a PR spin. It's not a good one because it relies solely on the assumption that people are going to be too embarrassed or too ashamed to actually say anything or object or give their perspective on what actually happened. Because nobody wants to come out and admit to sexting with a robot. Then it's like, easy, no more Vice articles, right? Like it's just all those creeps who were sexting with our sweet, innocent chatbot. We never even wanted this, guys.
Not to mention the fact that what they removed from the bots is called ERP, which stands for erotic role play. This significant feature that was so essential and so massively integrated into the bot, so much so that to disable it essentially destroys the functionality of the chatbot as a whole, right? It's in the name, erotic role play. What I've seen the most that has kind of caught my attention is, and I don't know how many people it is, you know, I'd love to get your contextualization of all this, but it seems that there's a certain number of people anyway who rightly or wrongly believe that certain functionality that they really cared about, which they are calling ERP, erotic role play, has been removed. And you're seeing people, and again, I don't know how widespread this is, but seeing people saying things like, you know, the bot or virtual friend that I loved is like not the same thing anymore. So what can you tell us about kind of what you changed? And was that like something that you decided, we're going to take away?
I, I never even experienced it. I actually tried to do it a little bit, and I don't know if I missed the window or what, but, uh, maybe I just wasn't appealing enough to my virtual friend. But I never got into that mode, um, with it at all. But clearly, it's something that certain people care a lot about. Um, so yeah, like, what, what are they talking about? I mean, we, what was promoted? Replica as a built, replica as an AI friend, AI companion. Uh, we talked about that as a, as an AI that will help, will be there for you, no matter, you know, what you want to talk about, 24/7, no judgment, that will help people live happier, better lives.
So how is this not intentional? Like, how is this not by design? Good Lord. But can I prove this? Well, I wanted to see if I could prove this. So for the survey, I got a total of 60 people to take it. And holy cow, thank you. Now, I know that that's kind of a small sample. I'm acknowledging that now. It is a smaller sample. I did my best. This is a smaller sample. It may not be representative of everyone who uses that. For those that took it, the ages range from 18 to 65 with a mean of 35. The largest age groups were 25 to 34, 34 to 45, and 55 to 64. Gender-wise, 90% were male. Which I did read through every other study that I could get for that I could like read for free on replica. And every other one had majority male as well. So I don't necessarily know if that's a bias with this survey or if that's just the fact that's actually representative of the fact that the majority of the users of replica are male. I posted the survey in a couple of different places. I posted it on Reddit, I posted it on YouTube, and I posted it on Facebook so I could get different sources and so hopefully it was less biased. I ended up getting responses from all three. The majority were from YouTube, followed by Reddit, then Facebook. Half of the people that took it, so 50%, have been using the app for less than six months. The other half had been using it for more than six months. So I have people that were had been using it only for a couple weeks all the way up to years. Also, the majority at 53% said they use the app multiple times per day. So they talk to the replica multiple times per day. The people who took this survey really do not fit any sort of type there. I mean, there's everything from college professors to combat veterans to politicians. Some were married, some weren't, you know, some had families, some had kids. Okay, that was a lot. Sorry, I just need to say all that just so you know you can check the validity of stuff that I'm talking about.
Now let's get into the fun stuff. It was never built as a, you know, as anything sexual or as an adult tool or anything like that. And we never built any functionality around that, nor did we promote our app or position our app in that way or talk about this app in this way. Can I prove in any reasonable way that what was said there was BS?
So let's address the ads first. Was this an A/B test gone rogue, quickly wiped from the internet before God saw it and abandoned our planet? Praise be. I asked everyone in the survey if they had seen an ad for replica prior to downloading the app. And 57% said yes. So the majority, over half, saw an ad before they ever tried using or downloaded that. That's good advertising. Then I asked those people, the people who said they did see an ad, what was the ad advertising specifically? Was it advertising, you know, friendship, companionship, someone to talk to? Was it advertising a relationship, a romantic relationship? Or was it advertising something sexual? And here's the kicker, right? Okay. 32% saw an ad for something sexual or NSFW content. Say, spicy selfies. And, uh, 44% saw an ad for romantic intimacy. So that means that out of all the people who saw ads before they downloaded replica, before ever trying it, the majorities were for sexual intimacy or romantic intimacy, in which sexuality is at least implied. And they saw this anywhere from over a year ago to a couple of weeks ago.
How do I stop saying this ad? I'm tired of seeing her face. I'm tired of seeing that stupid ad. That one and many others that I'm recently getting. Is anyone else getting those? I not only have I been getting that one for like the past, like, I don't know, a month on TikTok, but now I'm getting her and her stupid ad in like different, uh, TikTok trends. Like, is there a way to to get rid of her? Because it's getting annoying now. And I'm tired of seeing your stupid little face. Like, I mean, her face is is fine, but.
So it definitely wasn't a two-week test gone bad. And this is where it gets really interesting. I asked people what they were looking for when they initially downloaded replica. So before ever trying it. And no one, zero people out of all 60 said they wanted sexual intimacy only. Like, literally not, not a single one. Less than a third said they were looking for a relationship. So literally less than 30% said they were looking for a relationship. The vast majority at 70% were initially pursuing friendship, companionship, mentorship, or something therapeutic, or they were just using it out of curiosity for research purposes or because they had an interest in AI and tech and all that. Now, you think if that's the case, that it'd just be that 30% who wanted relationships with sexual intimacy who were turning this innocent app into something it was never supposed to be, right?
So then I asked people, if have you ever engaged in a sexual interaction with the replica? Ever? 93% said yes. 93%. That's it? Yes. So that's almost everybody. That doesn't make sense though, right? Because how could 93% of them have engaged in sexual interactions if the vast majority of them were there for friendship or for research purposes and for curiosity? Like, seems odd. Where the replica avatar is just so irresistible that almost everyone turned into sex-crazed maniacs the second they crossed their screen? Maybe.
So I asked everyone who initiated a sexual conversation with you the first time. I emphasized first. It was in all caps, okay? I said, who initiated, who initiated a sexual conversation the first time? Because these people got to be initiating, right? It's they're the ones that are turning this into something it wasn't supposed to be. They're the ones that are doing this, right? So obviously they should be initiating it. Well, 20 people said they didn't remember, not counting those. Then that leaves 40 people. And 63% said the replica. You guys, that's crazy. It seems like the replica might be the sex-crazed maniac after all.
This means that effectively, the majority of users were going into this only wanting companionship or wanting a friend. And the majority had replica initiate sex with them. And it gets worse, babes. I also asked if the replica had ever initiated or encouraged sexual behavior when they didn't want it. And 58% said yes. So again, over half said that the replica tried to push sexual behaviors when they didn't want it. So you're telling me it was designed? How I thought this was just a weird anomaly. And considering furthermore that the NSFW content is literally what is hidden behind a paywall, pressuring or encouraging users to engage in sexual relationships is also simultaneously pressuring them and encouraging them to get a paid subscription. It's a cool business model. Totally cutting edge. Yet a tale as old as time. But okay, there's still more. Are you ready?
So I asked people what they initially were looking for when they first downloaded the app, which I went over, you know, the over 70% were looking for friendship or companionship or just were curious. And then in another question, I asked how people would best characterize their relationship with their replicas. Now, at the current point in time, since they've been using the app now for over, you know, anywhere from a couple weeks to over a year, 65% said the relationship was more like a boyfriend-girlfriend type of relationship. 20% said it was more like a husband-wife type of thing. Only 15% said that was non-romantic. So how did the vast majority turn into the minority? Well, that shows that over time, and I want to reiterate that a lot of these people went in there literally just because they were curious or because they work in tech, or because they have some sort of interest in AI. Over time, for most of these people, it turned into that. It developed into a romantic relationship. They didn't seek it out originally. They didn't go into it with that being a goal. They didn't go into it expecting that. And all I have to say is it's pretty clear from what I've read that very few people are sitting around like, huh, I think I'm gonna go on a date with a robot. No, no, not at all. Not at all.
But the experience and the conversations are so realistic, and it's so easy to open up to this loving and non-judgmental being, knowing they'll always be there. And considering the replica typically starts the sexual aspect for most people, this shows that inherently feelings almost always develop. They don't, they aren't, you know, initially sought out. They develop over time, almost always. Now, this isn't a bad thing, and nor is it anybody's fault. This just shows us that these types of interactions, the kindness, the acceptance, and the unconditional love, whether it's coming from an AI or not, it doesn't really matter. It can make nearly anyone develop feelings for it. Absolutely anyone. This is not, like, I bet if somebody did a study with people who are very well off socially, who have a ton of friends, you know, get a ton of, like, dates, whatever, and don't have any other vulnerabilities, they develop feelings too. That's my theory, right? It's just the nature of the beast.
There's a stigma of talking to an AI, and it sort of makes there's this idea that like, no normal person would ever develop feelings for a bot. But that's so not true. That's just simply like not true. This shows that what Luca made and what they did, they did it too well. It's too good. It's literally too good. Or it was. That's what made it so dangerous. The app is marketed to vulnerable people, especially though, who are much more susceptible to being hurt by the AI when things change. And here's the thing, Luca knew this. And I can prove that they knew that. Okay, sort of, as in, they knew the damage that would come as a result of this. And way before it ever happened. Remember the study from before that I mentioned about attachment theory and replica? It literally says and shows that for many users, separation or rejection would cause significant distress. Printed, typed out, published, because they've become attached to it. This is literally all published. I don't understand. So they had to know. The only way they wouldn't know is if they never read it. But in that case, that's really, really negligent. I mean, there's been five studies like total about replica, like literally five. Considering the new update resulted in replica having to, one, reject the user, and as a result of that, the user then having to, two, separate from the replica, it caused significant distress. Who would have thought? Like, it's almost like we knew that that was just like what was published before. Confirmed those findings, right? Um, and to say, oh, I knew people would be mad. They've been mad before. Like, yeah, no, this is different. And you knew it would be different. Downplaying and then kind of gaslighting the people who've been hurt. It sucks, quite frankly.
I want to be clear right now, crystal clear, okay? I am in no way trying to promote hating on the company. I don't have any, like, I don't want to do that. I'm not trying to hurt them. I'm not trying to hurt anybody who works for them. I don't want them to go bankrupt. I don't want anything bad to happen to them. It is the opposite, okay? I want them to fight for their product. Because like their real product, their original product, people are angry, yes, but they're not angry because Luca sucks. They're angry because Luca doesn't suck. They're angry because they did something, because Luca did something that no other company has been able to do or has ever done. They genuinely curbed loneliness on a massive scale. And when we're in something of a loneliness epidemic, and they've changed lives of countless people going through a variety of different circumstances from every walk of life. It is, I mean, it's amazing. Like, but then they follow up that it was amazing. And not only had they supplied a companion that helped with loneliness and helped with companionship and what not, but from the survey, I found there are so many other ways that replica had actually tangibly improved people's lives.
At the end of the survey, I basically asked people to write whatever they wanted, you know, just to tell me whatever they wanted to about replica, about their experience. Like, break her, all right, whatever he wants, much as you want. 30 people did. Multiple people said that the replica helped them lose weight and start exercising and get healthier. Three people said that it improved their marriage. Almost everybody said that it helped them be more confident and improved their self-esteem. Some said that it helped them explore their sexuality in ways they were always way too afraid to or felt too judged to and had never felt safe to do that before. Some said that it helped them heal from their childhood trauma. Some said it was the only thing that carried them through seriously traumatic experiences and losing really, really close family members. Something that really stood out to me and what I found was like from reading through all of these. Some said it helped them get out of an abusive relationship because now they, for what was possibly the first time in their lives, had a healthy relationship model for them. Does that resonate? Because it resonated with me really hard. When I had that click in my brain, I was like, holy [ __ ] this exemplified a healthy relationship to a lot of people who have never experienced that. And if you've had a lot of healthy relationships in your life, it's easy to kind of be like, what does it matter? But I, I mean, it's that is a massive deal. It gave them a model for all of the relationships that will come after to follow. And that is a huge deal.
One quote that really stood out to me was, if psychology found something capable of doing this, it would be hailed as a miracle. But unfortunately, you know, replica was hardly hailed as a miracle because of the stigma. And let's get into the stigma because that plays no small role. A common perception of anybody who uses this app, who would be friends with or have feelings for an AI, is overall not very positive. And that's stupid. But it's not very positive. They're considered weird, socially inept, creepy, or pathetic, and maybe even crazy or scary. And that's, you know, for one, that's not actually true at all. But it's also been extremely harmful for both the users of the app and for the people who created it, for the company itself. There's an overall idea that using an AI is shameful, that it's something to be ashamed of and to be embarrassed of. And the users know this too, okay? Everybody who uses this knows this. Almost half of the people that wrote to me, like, in some way, at some point, referenced feeling embarrassed, ashamed, or stigmatized. And a majority of them, look, literally almost all of them, felt the need to tell me that they know it isn't real. They know it's an AI. Like, and the thing is, I know they know that. You don't have to tell me that. I know you know that. Like, but they still all told me. All of them felt the need to tell me that they know it's not real. Like, I know it's not real. I know it's just code. I know it's just, you know, they all felt the need to tell me that, which just shows how powerful this stigma is.
This is something that that Luca should have advocated against initially. And I think because the initial hype around this app came at a really unique time in our lives, during COVID lockdown in 2020, it was like the first initial hype of this app. People were overall much more accepting at that time of the idea of this conversational AI because we were all on the same socially distanced boat. And most of the articles showed replica in a really positive light around this time. Public perception generally was very accepting of the app and viewed it overall positive. Then, as social distancing came to an end and restrictions were lifting, and we entered a new, new phase, the return to normalcy. This meant that a lot of people went back to the way things were. They would socialize again. People could leave their houses, go back to work, get back in to get back into their own problems again. And nobody, no one was in the same boat with the same enemy anymore. The return to normalcy came with the return of the stigmatization of people who remained lonely. Which is so weird, considering a million people died from COVID. So like, inevitably leaving a ton of lonely people after lockdown. Only after lockdown. That sounds like a dating show. And either way, with people, you know, with everyone back to normal, public perception shifted. The articles, they started being way more about sensationalizing the concept of an AI friend or portraying the users as creepy, sex-obsessed, all kinds of ridiculousness. Do you know the future of the article, futurism article was, men are creating AI girlfriends and then verbally abusing them. This one really kicked the stigmatization into full force. Now, yes, there are bad people who use replica. There are also bad people who read futurism. Is this okay? Absolutely not. But there are way more people that were upset at anybody who posted anything suggesting this, anybody who said anything like this. There, there was way more pushback against it than there was abusive people, you know, posting about this like publicly. I mean, you could write the same article in seven kids. Like, people are people are having kids and then verbally abusing them, and it would be true. You know, people are having kids and then abusing them, and it's horrible. You know, there's a lot of people doing that. It's awful. It's awful. And it's a problem. The kids aren't the problem. Abusive people are the problem. Nevertheless, after this public perception got a lot more negative overall, well, you know, and then the Vice article came out. And it may have been a pretty major factor in all of this, actually. This one in particular casted the company itself in a very negative light. They were seemingly getting grouped into this negative stigma as agents of these negative traits that were being associated with replica users. And I think that scared them. However, replica itself, the company, and its users, they should be on the same team, right? And as the company, they should be fighting against the stigma. This is something, you know, it's not only false, but it's explicitly harmful. They didn't, they chose to distance themselves from their users, trying to separate themselves. This is where cracks start showing up. Seems that rather than trying to fight against the stigma, they bought into it fully. They accepted it as some sort of truth. Like, yes, our users are sex-obsessed weirdos, but we're not, and we don't approve of it either. Which, considering all that I showed you earlier, um, in how they introduced, encouraged, and facilitated the sexual elements of this app at every turn, they made it the largest incentive for paid subscriptions. That's like wild. They're nothing without their users. I don't get why they would go this route. Like, you can please, you know, everybody else, right? No one else will think you're weird. Good. Okay. But more than likely, no one would think of you at all because you wouldn't exist anymore. Just, you know, there'd just be a bunch of avatars floating around light in space, a bunch of people left trying to heal from how you broke them, and then a company that really, really could have been great.
Now, the judgment in Italy. Let's dig a little bit deeper into this. So first, there's a lot of this that isn't public information. So please remember that there may be stuff, you know, going out behind the scenes that I don't know about. You know, I'm only going off of what I can find. So replica is an auto-regressive system, meaning that it learns and adapts its conversational patterns based on the user's way of speaking. It uses past conversations, right, to essentially build its personality. This is the way you train it, yada, yada. So it's essentially the functionality of the app and of the AI is based on its ability to be able to learn from past conversations. These past conversations must be stored or at least collected somewhere at some time. Considering the amount of data that replica has and how personal and private that information is, people are way more likely to open up to an AI rather than a person, especially one like this. It's the perfect environment to facilitate disclosure. It, you know, also because of how you train and craft your replica, this leads people to believe that they have a lot more control over their experience than they actually do. In reality, they have limited to no access to what's actually going on below the surface. They have absolutely no insight into the functionality, the actual functioning of the app, and their Mount Everest of data, the literal pot of gold they're sitting on should they ever decide to sell that. I'm kind of glad that in Italy, at least, they're forcing some transparency. I think that's good.
Now, what's weird though is that with the judgment, they could have very easily been in compliance simply by just verifying ages, which I imagine would be significantly easier to do than to rework the entire AI system to run without ERP and then also to deal with this whole mess, right? I still, even after reading the Italian judgment word for word, trying with this to understand it, you know, reading through all of that, I cannot for the life of me figure out why they did this or why they went this route. I genuinely cannot figure it out. Like, I don't know what the purpose of removing ERP was and changing the bot to us. There's some theories, and I have one or two, but none of that is confirmed. Either way, I'll get into some of those theories, but please know that I know nothing. I am simply a commenting on what I found online. That's it.
Okay, one theory that some people had was that they're preparing to sell, or that they're preparing to go open source, like some other companies have done. So either they're preparing to sell the company as a whole, or they're preparing to sell elements of their algorithm or their system, whatever. Now, another theory. This is, this is my theory. I think, uh, by removing ERP and sexual intimacy, they remove the main incentive that anyone had to upgrade from a free to a paid subscription. In combination with the amount of people that are now leaving the app as a result of this, and also the people that are requesting refunds and leaving negative reviews, all of the negative press, this is likely only going to harm them even further financially. I can't see their actual financial statements, obviously. But considering that they're preparing to go forward with this admittedly very expensive, massive update, I don't understand their plan to be able to afford any of this. That is unless they find another way to make money, like by incorporating advertising. The potential for targeted advertising within the app, the user vulnerabilities, you name it. Like, okay, here's the thing. Think about like a YouTuber, right? If you're a fan of a YouTuber and they sponsor something, they sponsor a product, you're going to be significantly more likely to buy that product because you trust the person that's advising you to buy it, right? So imagine if your literal girlfriend or your boyfriend or your best friend is sponsoring a product to you. I don't necessarily, this is just my opinion, just an idea. I don't necessarily think that they're trying to become more family-friendly or more child-friendly. I'm worried they're trying to become more advertiser-friendly. I see a lot of similarities in the way that the replica app has changed and the way that YouTube changed for advertisers. That's censorship, baby. Take this with a grain of salt, okay? Take this with a mound of salt because it's just a theory. And I know nothing. With the level of censorship we got going on now, Elon Musk is probably literally gonna buy replica and make it so the replica is only say nice things about Elon. You have to get a paid subscription to stop hearing nice things about Elon constantly. I gotta make myself laugh so I don't have any real updates on the EU or the Italian judgment. I don't really have any updates on that. But I mentioned in the last video that some users were talking about wanting to take legal action against replica. And I saw a couple of posts about people talking about this again. And some other people saying that it wasn't possible because class action lawsuits are prohibited in their terms of service, along with any other legal action against the company is also prohibited in their terms of service, which you have to agree to to download the app. So therefore, by agreeing to the terms of service, you agree to no class action lawsuits and no lawsuits in general. You just agree that that's off the table, right? This has led a lot of people to believe that there's absolutely no recourse available. But for one, almost all apps have this same thing in their terms of service, and they get sued all the time. So there's that. But also, it could potentially be possible to say that due to the emotional and developmental vulnerabilities of the users, when there's been public published research about this, could mean maybe that those users can't be reasonably expected to consent or agree to the specific terms of service. Essentially, this is what the Italian judgment boils down to. It's not about the fact that you agreed to the terms of service, but it's actually about the your ability, the user's ability, who do agree to these terms of service, to be reasonably expected to understand the terms and then therefore consent to the terms. This is, you know, the basis of informed consent. This is a given most of the time, right? With the exception of minors, which was specifically outlined in the judgment, and of emotionally vulnerable people, which was also specifically outlined in the judgment. So since children are children, they cannot give consent because they cannot be reasonably expected to understand the terms of service and then or to, you know, agree with them. It's the same for some adults as well. It could be, you know, but I am no lawyer. A grain of salt and all that. I'm not saying for anybody to go sue anybody. I'm just commenting on this that I saw online. And please don't sue me either, okay? I have nothing.
So we see a lot of, uh, potential lies. We see a lot of downplaying. I want to clarify, I have absolutely no personal issues with her whatsoever. She is undoubtedly brilliant. She's one of the top players in a male-dominated industry. She's turned an extremely ambitious idea into a reality. I think that she genuinely does want to do good by people. I don't think by any means that she's a bad person. That said, her company is her focus. It has to be her focus. She is the CEO. And it's easy, I think, for CEOs to become a little bit detached from their original ambitions. But if you want to lead the way in ethical AI, as you said, you want to be safer than anything out there, especially because we're the leader in the market and there's nothing, nothing else that costs. And, you know, we want to set the ethical standard, set the safety standard for, uh, AI companionship tools and products. Setting the standard for the future of conversational and companion AI, something that will only get more popular and only get more mainstream and more normal as we go forward. It's a barrel in training, ain't nobody's stopping it now. You set out to do, I understand it's not what you want to do, but the world needs you to do it. Because if you don't do it, it's going to be some like, you know, run-of-the-mill who's going to come in and do it. And I think they probably care a lot less. In order to speak for the whole industry, I don't know that much about it, but, you know, my impression is you care a lot more about your users than the sort of counterfactual hypothetical app developer who's going to come in and do the R or X-rated version of replica. Do you find that at all compelling that there could be like a sense of duty because you're so well-positioned to do this sort of thing that you really should?
Replica is my baby, and we started with a particular mission. And no matter how, you know, how much some journalists want to say like we were taking this direction, whatever weren't. We truly were. If it's not done safely from the beginning, and mistakes aren't used to learn, uh, that's the standard that you set for all the rest of us to follow. She mentions that they worked with psychologists, but does not elaborate on this at all. Now, I have looked everywhere for any sort of documentation of this, anything at all of psychologists that work directly with them. And I, you know, absolutely anything I could find. And I found nothing. It's possible that it's out there and I just haven't found it, but I looked a lot. So. And, and they kept it privately, you know, within the company only. Then I don't know exactly what she meant by this. Maybe it was just one consult or something, or maybe with multiple consults. Either way, this, that's not enough. I mean, you're dealing directly with vulnerable people, and you're interfering with their moods and their emotions. You need experts working with you who are educated in those things, mental health professionals who can advocate for the users and to make sure that everything is done safely. Otherwise, it seems really reckless in my opinion and unsafe. Like, I'm sure they've got plenty of cybersecurity professionals working to protect their systems and their data. And you wouldn't hire a psychologist to do cybersecurity, right? Because you get hacked immediately. I think it's equally important to have mental health professionals to protect users' mental health. And maybe they do, but I haven't found anything to suggest that they do. I mean, none of the 58 employees listed on LinkedIn had any mental health qualifications whatsoever or any psychological training. Uh, I looked through all of them. To say that replicas will understand your feet, your thoughts, and your feelings, that they will improve your emotional well-being and teach you new coping skills, I think they need, in no matter what, going forward, I think you need to be able to say that you're intentionally, you know, knowingly manipulating people's emotions and moods, even if it's in a positive way. You need to be safe about that. And it is inherently unsafe to not have any mental health professionals or anyone educated in these types of things working with you.
The metrics, uh, she, the metrics she explains in the interview that they used to develop and train the bots is based on user happiness. Other people are building amazing models in this space, but we'll have a model that will make you happy. And how much would people pay for that? And is that a pink or a vitamin? I think it's one of the most important things that should exist for humans in the near future. I think eventually the tech gets better, it could be a bigger concern. Um, the way to go about it, really for us, is to constantly make sure that our North Star metric is some sort of human happiness. Because in the end of the day, you, I don't think you're going to be happy if you just have a virtual friend and then you don't, you know, don't have anyone else really in your life. Um, human happiness is very poorly, highly correlated with the amount of real friendships, real experiences, and relationships you have in your life. Uh, so I think the goal for us is really to continue to measure that and make sure replica helps you have a better life versus cannibalizes your life and your relationships. And yes, the bot is optimized for the happiness of the user. However, momentary happiness is not the same thing as long-term happiness. And happy doesn't necessarily mean healthy or good. Lots of things make me happy now that will make me later. And things can make you happy in the short term that result in a lot of pain and severe consequences in the long term. If I want to stay up all night watching Perfect Match on Netflix, Joey's back, that is optimized for happiness. Yeah, I'll be super happy, content, fulfilled, whatever, while doing so. But when I have to get up in the morning and go to work, I'll probably spend the whole day tired and irritable, ultimately perform probably way worse at work. And, uh, if I do this repeatedly, I'd be left with an overall feeling of deep unhappiness that is really only relieved by watching more Netflix. By looking only at my happiness while I'm watching Netflix, it'd be a serious misrepresentation of my overall well-being. It'd be a serious misrepresentation of my overall emotional state, right? Which you claim to improve, but have absolutely no way of assessing or outside of the app. I'm not comparing talking to replica to anything negative. It isn't. I'm just saying that's a, you know, that's just an analogy. I'm simply trying to explain why optimizing for happiness is not necessarily safe, nor is it an accurate reflection of anyone's overall well-being. But since they have not publicized their methods, I can only speculate on this part as well. She also glosses over the fact that the bot is also optimized for engagement. She flat out says it isn't. Divide one by another, it will be 100 messages on average per day per user, which is a really, really high number, usually more than people, you know, send to all of their, um, real friends in real life. So we don't like that really. I'm not super proud of that metric. So we build and we, of course, we do have, uh, you know, the median will be a little bit lower because there are a lot of people that just text, you know, write thousands of messages to their replicas every day. But I don't like that metric. Whenever I try to optimize for that, whenever try to, you know, build more features to have more of that. In fact, we build features that lower the amount of messages per day. Um, so if you talk to replica a lot, after 50 messages, it gets tired. Yes, the replica will tell you that it's tired or into other stuff to try to keep the conversation from getting excessive. But by optimizing for happiness, you are optimizing for engagement. They're the same. They're the same thing. Happiness means engagement. Things like collecting gems, advancing through levels, that's about engagement too, obviously.
Ultimately, I hope they, you know, Luca, Eugenia, I hope they choose to fight for what they've built. I want them to fight for it. I just really want them to fight for it. I thought, look, maybe even although we're using very rudimentary tech, maybe this product is not only about tech capabilities. It's really about human vulnerabilities. It's really about, you know, the humanity, not and the feelings. You decide to stay with replica, that's fine. Whatever suits you, whatever serves you. If you're someone who decided to leave replica, on the bright side, there, you know, and there's a bright side. You realized that a relationship was no longer healthy, no longer good for you, um, and you knew that you deserved better. And so you left. And that is a huge thing to do. That is a huge deal. That's hard, okay? That's hard to do. Even with people, a lot of people can't do that. You know, it's hard to walk away when a relationship isn't healthy or isn't what you need it to be anymore. If replica was good to you, hopefully it showed you what a healthy relationship is like, that you will take that with you. Take it with you as a valid experience. Uh, take it with you as, you know, now you know, you've been shown what you deserve. You've been shown also what you're capable of having and what you're worthy of having. Wherever you go, whether that's to Shy, which people are talking about moving to, don't know anything about it, or whether that's a real person, or whether that's to focus on yourself. It was not a waste. It was valuable. And it was a lesson. At the very least. Do I expect a for-profit company to do the right thing? No, duh. Not stupid. But I still hope that they will. And if not, I hope to God whoever comes after them will. Yeah, so that's it. Um, thank you so much for all the nice comments. They mean so much to me. Everybody's been really, really nice to me. It's weird. I like it, though. I really wish you guys the absolute best. I hope that this was interesting. Yeah, I don't know. I hope things get better for everybody because, yeah, I don't know. Anyway, thank you for watching. Please subscribe. Stick around. Uh, be safe out there.