Meet Your Future Self with Pattie Maes and Pat Pataranutaporn

What if you could use AI to see what your life would be like when you are 60, or better understand how you got there? Today on The Future of You I’m talking to the creators of an interactive experience that will allow you to do just that.

In this episode I speak with Germeshausen Professor Pattie Maes and technologist and researcher Pat Pataranutaporn from MIT Media Lab. Pattie and Pat are working on "Future You", a project using AI to create our own digital twin.

We talk about human-AI interaction, human attachment to machine conversation, and what questions we might pose to our future selves when given the opportunity.

Try Future You from MIT Media Lab https://futureyou.media.mit.edu/

Tracey's book 'The Future of You: Can Your Identity Survive 21st Century Technology?' available in the UK (https://bit.ly/44ObTha) and US (https://bit.ly/3OlDxgk)

The Future of You was named Best Technology Podcast at the Independent Podcast Awards 2023.

Find Tracey at https://www.futuremade.group/abouttracey

Explore more on The Future of You at https://www.futuremade.group/the-future-of-you

Transcript

SPEAKER_00 0:19

Welcome to the future of you. Today I'm talking with the Fluid Interfaces team at MIT Media Lab, specifically Patty May and Pat Pat Taranou Tapon. Professor Patty May is a bit of a legend for anyone interested in human machine interfaces and software agents. The Belgian-born computer scientist created a music recommendation agent that connected people with shared tastes and interests, Firefly, as far back as the mid-1990s. She's worked at the lab for over 30 years now, telling me that her interest has always been human augmentation with digital technology. Pat Pat Taranu Tapon is a PhD student. Well, he's just completed his PhD successfully, working in the Fluid Interfaces team. His research work focuses on trying to understand the art and science of human-AI interaction and applying that to the building of better AI systems for human flourishing. And both of them have been creating this, well, interactive digital twin as a system for self-reflection and personal growth called Future You. I've been following their work for a while as it dovetails so perfectly with the Future of You, and so it was my great pleasure to get to speak to both of them about Future You and more generally about AI human interaction, about human attachment with machine communication, and of course where all this is going next. This episode really is at the heart of the belief system of the future of you and the notion of technology of the self, since it brings to life something I predicted in the book, a simulated version of a future you. I'm so happy to bring this chat to you. I really hope you enjoy it. Patty and Pat, welcome to the Future of You. Thank you for joining me. Thanks for having us. Not at all. I was just saying to Pat that um how could we not talk about the Future U on the future of you? Um so I wonder if you could explain to our listeners what Future You is, why you've done it, and just give us a little bit of a flavor about what problem you think it solves or how it adds to the human experience?

SPEAKER_01 2:37

Yeah, so uh the Future U is a project uh we are currently uh working on at MIT Media Lab Um that used AI um to simulate a version of you in the future, you know, that's what the name implied, so that you can have a conversation with your future self. Um right now we are targeting kind of a younger audience and allowing them to talk to a 60-year-old version of their uh future self. Um I got the inspiration for this project uh you know a very long time ago when I was a kid, watching this cartoon from Japan. Uh the cartoon has like a robot cat from the 22nd century that comes to help this boy who doesn't like to go to school by taking him on a time machine to see his future self. And you know, he you know realized that if he doesn't pay attention to his you know school and grade, then he will end up not doing well in the future, and that kind of profoundly changed the protagonist. And I thought it was such a powerful use of technology that you know help you realize you know the consequence of your action and help you think in the long term. So um that sort of stuck with me for a long time. Uh, and then when uh we have this AI technology and the ability to actually you know use uh technology like deepfake to actually you know change people's age or generative AI to generate conversation. I think that we can actually create that kind of virtual time machine. You know, we don't have a real time machine yet, but we can make this simulation so that you can actually have this uh virtual session with the future self. And um we work with uh Professor Hal Hirschfield, who is an expert on future self continuity, he's a psychologist from UCLA to evaluate what happened when you actually uh talk to this uh virtual self from the future. And um in our recent study um that just got accepted to IEEE Um Frontier Education, actually, we found that this type of intervention can reduce uh people's anxiety about the future and make them better connect with their future self. Like they think that they can vividly uh imagine what their future self might look like, they can better take that sort of perspective and hopefully have a decision that you know geared toward that future, not just the immediate situation that they are in right now.

SPEAKER_02 4:49

Yeah, a lot of us, um not just young people, but especially young people maybe have a hard time with taking actions that ultimately will contribute to their long-term well-being. And what we do with this application is that we really make that possible future a little bit more tangible, or we can make multiple possible futures of yourself more tangible, more visible to people, and let them really talk with this future version, possible future version, I should say, of themselves. And just making this more tangible, more experiential, less abstract makes it also more effective in terms of long distance thinking or long-term thinking and and having people or influencing people to take actions that ultimately lead to long-term goals that they may have.

SPEAKER_00 5:50

I'd like to personally thank you because as a futurist, I spend all of my time trying to get people to think further ahead than you know the next quarter or the next week. And for some people it's very, very difficult for them to envision the future. I mean, particularly if you're in an organization, you're trying to envision the future of a market or a future consumer. But actually, sometimes it's even more difficult for people to envision their future self. So this I mean, I think this is I'm sure it's early days with it, and we can talk about where it where it might go next. But uh I am fascinated by this idea of self-continuity, as you have expressed it in a lot of your papers. I wonder if you could both just talk me and listeners who might not be familiar with that concept through self-continuity, and as you noted, you know, some of the beneficial effects that if you can get people to buy into self-continuity or almost activate it in themselves, what it can bring.

SPEAKER_01 6:46

No, that's a great question. The concept of future self-continuity, I think, was proposed um as a measurement to measure how well uh can you vividly uh imagine or connect uh with your future self. And this measurement, you know, have been used in actually many experiments, actually, you know, and there are many intuitions that have been developed uh to help with this. But you know, study have shown that people with high degree of future self-continuity, they tend to have better mental health because they can, you know, see that the problem or the challenge that they're facing now will not be forever. They can see a version of their future self that you know get out of that challenge, for example. Um they tend to have uh better financial planning because you don't just plan for the immediate gain, but a long-term gain or long-term benefit. Um, and also in school, there was a study that showed that students that have a higher uh degree of future self continuity tend to perform very well because you know they can think about okay, what are they gonna use the education for, right? Not just coming to class and just having fun. Uh I mean that's the important part, but if you can think in the long term, then you can see how your life kind of uh leads up to that. And I think the challenge is that you know most people don't have the this high score on future self-continuity. So we need to develop intervention or system um that can help with this. And I think our work, um, you know, the the future you project is a technological version of it, but that auto approach as well that can be combined or used with this. I think it's a really, really important concept. And I think, you know, similar to the name of this podcast, right? It's not just trying to make abstract concept, it's trying to address like what happened to you particularly. And I think that's uh the power of future self-continuity is that is emphasized on the individual uh experience and and how that person can navigate that uh future. Yeah.

SPEAKER_02 8:34

Yeah, I think being able to sort of imagine yourself in the future requires a good imagination, and often we are limited in terms of um what we can imagine for ourselves. Maybe if somebody grows up in a family where both parents say uh if they have two parents, have accomplished a lot, and so they can envision sort of this future version of themselves being accomplished professionals or something. But if somebody grows up in a family where maybe the parents haven't uh achieved much or are not necessarily a good role model, it cannot be very difficult to imagine what possible futures exist, even and what those futures may be like. And I think our system really helps with making that possible future a little bit less abstract and making it more visible, literally, uh, so that uh that can help people see this opportunity or this possibility that they may have that they otherwise wouldn't be able to envision.

SPEAKER_00 9:48

One of the interesting things, I think there's lots of interesting layers to this, but is the the almost um counter trend around individuality. So so much of what's done on the internet or done digitally is about groups, the collective, community, what does the group think? I mean, there's always going to be an element of that because there's connectivity. But this idea that actually you're doing something for yourself, it's a bit like self-talk in therapy. Um, it's sort of going back to your instinct or your intuition or something that's deeper than what does the group around me think is best. I don't know if you can speak to that. I I think it's interesting because you knit together both the trend for the collective community, but also sort of individuality.

SPEAKER_01 10:30

Yeah, I think that's an interesting observation because you know the AI that we use to generate this future self backstory in a way is trained on you know a large data, right? It's contained information or profile of many people, but at the same time, we also try to personalize it. So, you know, it's not like you're just seeking an advice from you know uh an anonymous AI that have all this information. We present it as a version of you from the future. So, in a way, you know, um this is an empowering sort of narrative that you're not you're not just seeking an advice from a void of data, but you're actually getting a help from your own self, right? But just extrapolating or interpolating that from now, right? So maybe a version of you in the future that have more skill or you know, but but it's still based on you. So I think that's something really interesting, is that it's not about AI helping you, it's you helping yourself through the power of AI. I think that's what we're trying to create.

SPEAKER_02 11:27

Yeah, and I think that if we look at surveys, for example, of how uh these uh chatbots LLM uh systems are being used, we see in a lot of these surveys that uh one of the main usages is people just talking about themselves. And uh it's I don't know whether I would call it therapy, because of course a lot of this isn't even been tested or evaluated in any way, but it seems that there is a real need for people to explore sort of their own identity, their own possible self, and they are using these systems as um yeah, this mirror and this uh sounding board, really, to talk about all of the questions that they have about how they should live their lives, how they relate to other people. And it is possible to do that because LLMs collect all of this writing and have been trained on all of this writing by people expressing their concerns and the struggles and the things they are working through. So there's a very rich database there or amount of information, but it can be very personalized by these systems, and increasingly I think we will see that these commercial services as well will become more and more personalized. They will get to know you and have a long-term relationship with you, and uh will really act as a sort of a coach a little bit, or or yeah, again, this sounding board for just uh dealing with um all sorts of uh daily life uh questions.

SPEAKER_00 13:15

I totally agree. I mean, generative AI, there's a little bit of negativity around its advancements at the moment, a little bit of a backlash, I think. I don't know if you agree. Um you're in it, so you're the best people to talk to that. But I don't I don't really see that. I think it's just that people aren't necessarily being imaginative enough about where it's all going to go. So if I listen to what Eric Schmitz said the other day at Stanford or you know what Mustafa Suleiman said when he was in Inflection about how this is all tied to your tone of voice, your identity, it knows your objectives, and there's a fiduciary responsibility attached to it as well. You can see how it does become this version of you to go out into the world, well, more than one version potentially, to do things on your behalf. I mean, what's your point of view on the narratives around generative AI right now?

SPEAKER_01 13:58

I really like the quote from a technology philosopher, Marvin Skrandberg. He said that technology is neither good nor bad nor neutral. It's dependent on how you use it that actually, you know, dictate the outcome of technological use. Right? And I think, you know, our research group that have you know done a lot of study in this work, I think we have demonstrated that technology itself, you know, is really depending on how you use it, that will dictate whether it's gonna have a positive outcome or negative outcome. And generative AI by itself doesn't lead to you know improvement of people right away. You need to um you know really design, you know, even the system have ability to have like you know natural conversation with people, where would that conversation go? Or what would be um the way to use that kind of conversation to actually improve uh people's ability to think, like to learn, to think critically, and and things like that really require insight on how the human work, right? So um, you know, uh I think what we we try to do is really emphasize on the interaction. Um we put a lot of emphasis on human-AI interaction because the interaction would be actually the thing that dictates the outcome of this interaction. So um we try to understand like what kind of interaction leads to the outcome that we want, like, you know, in education context or in you know, even a future you project, right? We try to see, okay, what kind of interaction between the current self and the future self that could lead to this uh future self-continuity or like you know, better mental health. If you don't have that in mind and let the AI be the one you know creating agenda, it may not lead to the outcome that you want. So we should be the one setting up the agenda for where this interaction should go, not the technology setting up the agenda for us. I think the narrative today is mostly like, oh, we have this powerful AI, let's see how it's gonna take us. I think it should be the opposite where we know what technology can do and think about where humanity wants to go, and that's how we should use the technology for.

SPEAKER_02 15:50

Yeah, and going back to your question of the current backlash against AI, I think part of the issue is that AI was at least initially and still to a large extent introduced as a productivity tool, as something that would replace a lot of jobs or at least augment uh support people in their tasks and so on. And of course, because of the hallucinations that still happen and may well be unavoidable with the current models, um, it seems that as a productivity tool, there are limitations, and a lot of corporations don't want to take the risk of having these hallucinations really affect their work and their services that they offer and so on. But I think what we are again are starting to see is that they are possibly more useful as a personal uh life coach, a personal growth tool, as a companion, or at least a system that can help you with all sorts of uh life questions and emotional questions and relationship questions and so on. So it may well be that uh that is the killer app rather than productivity.

SPEAKER_00 17:11

Yeah, totally. I always laugh when people say about hallucinations. I mean, I know it's an important point, but as if humans don't lie. I think that shows how realistic they are. So for anybody that doesn't know, hasn't tried Future U, can you take us through a bit of the process? Because I know there's sort of a survey at the beginning, and obviously there's the image making of an older you. Explain to I know it's audio, but explain to people how it works.

SPEAKER_01 17:38

Yeah, so um right now um I think for the future you is you know, we we are doing the in the in the staff sort of launch phase um where people can actually go to the website uh futureu.media.mit.edu and actually try it out now. And you know, we are collecting feedback and want to you know improve the system and hopefully you know do a more experiment to really understand how it can help with uh people in this concept. Um but you know to your question, um the system I think have like three main components. The first component is that you know you come in, you answer some question, um, but we we have you know try to figure out what are the important questions that would lead to a realistic version of you know of the simulation. You know, this question range from like you know who you are, what you want to do, you know, your current behavior, your aspirations, you know, like especially for teenagers, like what career choice uh they might want to make when they grow up. And you know, all these are really important questions that we can then use to create backstory to actually generate the the future you character, right? And the second part is the image making, as you mentioned, we also create a version of you by kind of aging an image of you, the image that you upload to the system, so you can kind of see you know a version of you when you're 60 years old as a conversation partner. And then the third part is when you can actually get to interact with the future you. And we kind of you know, we we look at a lot of research on you know self-narrative and um research on you know what is the best way to engage people in this kind of exercise. And first, the future you will introduce him, her, or their self and talk about you know what kind of life experience that might happen, you know, between now and when you are 60 years old. So that gap try to kind of walk you through it and then offer you to actually have a conversation. Sometimes the future you will ask you questions to actually help you reflect and and make you think about you know why would you make certain choices versus the other choices. And I think the important part is of this research is that you know, this is not a prophecy. You know, what AI generates is not an ultimate uh future that would happen to you. It's just one of many possibilities. So we also encourage people to go back and change the answer to the early question and see how the new version of the future self changed based on that. So I think you know, we we kind of encourage people to use this as a reflective tool where you could go back and forth, kind of like taking a time machine to the future and come back to the to the now and and see how the now influenced the future and vice versa. I think that's the beauty of this uh system.

SPEAKER_00 20:12

So, what kind of questions have people asked the system? Have there been some surprising answers for people?

SPEAKER_01 20:18

Well, there are many surprising uh answers that we got. I think some people want to know who's gonna be the president in the future, which of course, future you cannot uh answer that. But for me personally, I think the question that that I still think about this day, actually, especially now I'm in Thailand with my family. One of the things that my future me told me uh was that when I'm 60 years old, my parents may not be around anymore. Like, and in that version of the future me, my parents would be like you know 90 or or or 100 or or something like that. And and my future me, the 60-year-old version of me, tell me to spend time with my family more and really kind of cherish this moment. After that, I call my mom almost every day. Uh, even I'm in Boston, like at least like trying to spend like five or ten minutes with her. I mean, that that fact is very obvious, right? Like, you know, you spend more time with your family when you grow older, your mom or your parents may not be around anymore. But when you hear that from your own self living in that life moment, even though it's a simulation, it could carry weight um that that really kind of make me think like, okay, if I'm at that age, need to tell or need to talk to someone about like what happened to my mom. Like, I I think it's kind of very emotional and very, very powerful.

SPEAKER_00 21:35

Yeah, once something's externalized, it is more powerful, like that, isn't it? Patty? Yeah.

SPEAKER_02 21:41

Um, so a lot of people, of course, I think use it to think about possible career paths. And what I find useful is that uh the system isn't just making that end goal a little bit more tangible and can show you what your life a day in uh Uh, in the life of say being a surgeon or a teacher or whatever would be, uh, but it can also show you or talk to you about the path that your future self took to get there. And that is important as well, because even if I can imagine my future self in some career, I may not see the path from where I am right now to that future state. Even if it's desirable, if I think it's desirable, I may not see, well, what do I do? What well what do I do next? Or what are the milestones towards that future? And so I find that to be very interesting as well. Not just experiencing this future and how I feel about it emotionally, sort of, but also uh learning more about the path from here to there.

SPEAKER_00 22:57

It's text-based at the minute, isn't it? Are there any plans to make it voice-based or perhaps even in the future, like embodied AI, so it does literally come to life?

SPEAKER_01 23:08

We did a version of that called Mekinoia, where you can see I actually that's based on my own data, see different versions of me that can actually, you know, react and and have facial expression. Um, it's somewhat terrifying a little bit as well. Um, because I think when you make a video, um, even with the current technology, the video generation uh technology tends to have like sort of the uncanny effect, especially if it's your own image or your own uh video, you know. I think it tends to be okay for like you know a video of like an actress or an actor because you don't know them very well. But when it's yourself, you can, you know, you know yourself very well, right? So the model tends to be a little bit off, and I think that could actually make it worse than having kind of something like a static image, right? So I think that's what we have been sort of looking into. But I think the the research that we are looking into now, we are really excited about, um, and this I think is partially inspired by the movie, uh, Everything Everywhere All At Once, as well. It's like, what happened if you show multiple versions of your future self at once, right? And especially right now we're looking in the context of career exploration. What if you help people, uh especially younger people, see multiple versions of their career path, right? You know, a version of them that become a doctor, an artist, an astronaut, a lawyer, a painter, or a futurist, even, right? Then then does that actually broaden uh their perspective of what they can become, right? So I think that's I think a really important uh uh role of AI is that it should expand our possibility, not diminish it. And I think that's what we are kind of excited about right now is to explore this kind of interface that allows you to see multiple versions of your future self at the same time and comparing that to um auto-conditioned and see what would be the benefit of that.

SPEAKER_02 24:56

Yeah, I have sort of mixed feelings about making it more realistic, like making it be video and voice and so on. Because as Pat mentioned, um, I mean, there is the uncanny valley issue, but also people start focusing on some superficial issues, like, oh, look at my belly when I'm 60 years old or something, rather than thinking of like what is do I like my life at 60 years old, uh, if this is the path that I take. So I think that's one issue. But one area where I personally think it might be interesting to do this is to show what the effects on others or on the world could be of your actions, uh, say 30 years forward. Um, as you know, people have a hard time sort of changing their behavior to be more, say, green or sustainable. And I have a suspicion that, and then this is one of the things we will try next as well, that if we try and show you sort of, well, here's what the world would look like if everybody was buying disposable plastic all the time, like buying disposable water bottles, etc. Here's what the world will look like 30 years from now, uh, versus here's what the world would look like if everybody used reusable containers for things rather than plastic. So making that future world again more tangible, more emotionally um perceivable and visceral, so that hopefully it it affects people at that emotional level where ultimately it changes their behaviour and it impacts their behaviour, is I think uh an interesting area to explore.

SPEAKER_00 26:50

Both those points you've made, which are different points I've been looking at for a while, like particularly the sort of multiplicity or plurality of identity or the idea of many selves. Because I think in the digital environment, I don't know what your point of view on this is, but we have to drop the standard of authenticity because it's more about profilicity, isn't it? About profiles, about arranging, rearranging characteristics, showing up in slightly different ways to fit the context, which um is not a very Western way of thinking about identity, but the digital environment is kind of forcing us to think about fluidity and plurality. I wrote a paper for the Journal of Future Studies on this, how we need to start thinking about the complex self, and actually that'll take us to more interesting places. Um but to build on that with Patty's point, that again is is you're thinking of yourself as having more agency, I think, once you think about the plurality, because you're starting to consider the consequences of all these things and having to make some sort of I don't know if they're trade-offs or decisions or go one route in the future, not the rather than just actually making decisions, stopping and making decisions rather than just carrying on as you did before out of habit, I suppose. Is it is that something you're trying to do? Almost get people to make better decisions, perhaps break some habits, form some new habits, healthier habits?

SPEAKER_01 28:09

Well, our group is called Fluid Interface, right?

SPEAKER_00 28:11

Yeah, yes, of course it is, yeah.

SPEAKER_02 28:14

But yeah, definitely that's part of uh what we're trying to do here. On the one hand, you can use this future you system to explore specific goals that you already may have and are wondering about and want to learn more about, but we can also use this type of system in a very different way and um basically design a whole experience uh that we hope will have certain consequences or a certain impact in terms of how people think about themselves and their behaviour and so on.

SPEAKER_00 28:52

I love it. One of the reasons I love it is because I do think people are rediscovering agency, their own agency through it in a in a world where lots of their time they feel like they don't have any agency or what they do won't make any difference, or maybe it's pointless. I wonder if we could just turn to the idea of human attachment. I wanted to ask you both, uh obviously through this project, as well as all of the experience that particularly Patty's got over all of the initiatives, even things like remembrance agents, all those sorts of things you've you've worked on in the past, but to ask both of you, where do you think we are with human attachment? Is it a good thing, a bad thing? Is there a spectrum on which we should think about human AI or human machine attachment? Love to get your point of view on that.

SPEAKER_01 29:38

Yeah, no, this is a really, really interesting question. And I really like your earlier point about the agency, right? I feel like, you know, all these, you know, concepts like attachment, uh reliance, and all of this, I think it's a factor that you can decide for or against. You can decide technology to be more addictive that people want to use it, or less addictive that people have higher levels of agency, like they can quit when they want to do it, right? Or or neutral as well, like you know, Photoshop, I don't think, is designed to actually make you kind of addicted to it or less addicted to it. It's kind of neutral, it's just a tool that's there, right? So I think when when we start to think of technology as a design challenge or a design question, then you don't just, you know, have it uh come as a full package, like AI addicted. No, AI could be designed in many ways, right? And the plurality, as you mentioned, I think is it can also apply to how we think about where the technology could go as well. I recently wrote an article uh with my colleague Robert Mahari for MIT Tech Review on um why we need to prepare for addictive intelligence, you know, AI equals addictive intelligence. And we say that, you know, AI is not addictive by nature, it can be designed many ways, but I think there's an incentive, economic incentive uh and psychological incentive right now to actually make this system that are very sticky. We have seen in you know uh many uh reports that when people talk to this sort of virtual companion, they use like pet names, for example, you know, like to kind of promote this. And this is was intentionally designed to make people come back to the session and use it more and more. But at the same time, if you kind of use it for great things like behavioral change, like helping people have like healthy habits by making AI systems that you know remind you in the right way or use the right kind of persuasive uh language or use things that you like or admire or care about. You know, we did a study on what happened when you learned from virtual Elon Musk. If you like virtual Elon Musk, um I don't know we can repeat that experiment today, but you know, that study showed that if you use a virtual character that people like or admire, even if they know that these characters are completely made up, they still, you know, uh have better experience or feel more motivated to engage with that character. Right? So this um sort of insight really show us that we can use this for good or for bad, depending on how we use it. Again, going back to you know uh Kranzberg point, that technology is not good or bad or neutral. It's really the intention of how we use technology that would lead to the certain outcome.

SPEAKER_02 32:09

Yeah, I think in in all of our work, our goal is to create tools and experiences that can help people become the person they would love to be, basically. Um, so it's really tools and experiences for self-growth, self-realization. I think a lot of us are just uh everybody is struggling with changing our behavior and becoming sort of more who we want to be. And we don't really have a lot of tools right now to help us with that. You can take pills to become say more attentive, or you can go talk to therapists, but they're expensive and not very available. While we have these tools that are with us, 24 by 7, that can be highly personalized, that can have a wealth of knowledge and information that they bring to bear on all of these interactions. So it's it's just an incredible opportunity, but also a huge challenge. And I think in terms of a as a design challenge, we always try to avoid creating systems that people become dependent on if possible, because dependency, if it's not needed, is never a good thing, I think. However, there are many situations where dependency is not avoidable. Uh, for example, you wear glasses, I see, and um, I mean, I guess you could get LASIK surgery, although that doesn't necessarily solve all uh all problems with eyesight, but you are dependent on the glasses, but that's not necessarily a bad thing because there is no other alternative. So I think we have to uh look at a situation and evaluate to what extent dependency is okay or not okay, and but minimize it if possible so that people have autonomy, have freedom.

SPEAKER_00 34:20

Yeah. I was thinking about something very specific, for example, and you may or may not want to comment on this, but you know, the Billy chatbot, you know, when Meta rolled out those chatbots and they'd taken celebrity voices and well, faces really, and so something like the Billy chatbot, which whenever I show that to anybody, they say, Well, well, it's Kendall Jenner, isn't it? It's Kendall Jenner's voice and it's Kendall Jenner's face, except for the fact that she introduces herself as Billy. And that's I think A is very confusing. It's not just complex, it's confusing. But also, I wonder, you know, how many teen girls, for example, are attached to that chatbot because of the way it's presenting ourselves. And I I I mean, I guessing that you would you would say, well, that's an example where you're building in some potential addiction that doesn't really need to be there, perhaps.

SPEAKER_01 35:08

I think if I remembered correctly, I think they they're they're rolling it back after this period of experiment, right? So um w I I hope they learn something. Um we we don't know, but yeah, I think that's a really interesting question, right? Like, you know, at one hand, if if this tool is used to kind of really promote interesting and and positive behavior, like, you know, how like with mental health or or something, I think that I could see uh the the the positive use case of that. But I think most of the time people that design these technology just like you know don't really think that far along. Like they just have this, okay, let's just talk to it. What what is the point? What is the ultimate goal of this? And I think one thing that that we need to recognize is that I think human, you know, right now, especially when you have very uh powerful technology, you know, I I don't think people want to be disrespected in that way, right? Like I think many, many, many, many uh people uh I think I have seen like don't want to talk to the AI because they feel like the AI is a cheap solution. Like you didn't want you don't want to watch a movie that was generated five seconds ago, and then you know, like you want to see something that you know had human touch and you feel that you know people making this are respecting you, that they're making something for you, right? You don't want to eat like a fast food of information. Um yeah, I I I think so so I think that people one of the reasons for backlash is that people feel that they are being disrespected a little bit to just see this AI-generated image that no one even looks at it. Like, why would I look at it if the people creating it doesn't even look at it, right? So I think that is one thing. Like if if this technology is used in a way that really um makes something that that you know is like a craft rather than a fast food, um, I don't know if the better metaphor for that, um, then I think it could lead to really, really powerful things. I've seen the conversation that happened in in Hollywood or or in you know uh filmmaking community, right, where they talk about like AI generated content versus you know human-generated content. And I don't think it's about it's about uh the quality, it's also about the the intention or like the the story behind the story that also matters, right? So um I think that's one of the reasons.

SPEAKER_00 37:17

I think that's really interesting. And then I suppose my question is really starting to get to just as we as we finish up here, your final thoughts, I guess, on it, but who deploys this? So you could be inventing it and uh creating it and researching it, but who then gets to deploy it? And if it is a big corporation who has you know some very commercial incentives, um, you know, can we ever safeguard against that? Because it feels like, and listening to you two, obviously it's incredibly thoughtful, it's well researched, it's human at the core. Um, how does one moving into the future try to ensure that that is actually what gets delivered and deployed to people in the future?

SPEAKER_02 37:58

Yeah, I think that is a very hard question. And to be honest, I don't know whether there is a foolproof solution for this. I think it's important that engineers and designers and researchers become more involved in thinking about all of these issues, the ethics of the questions that they are building, and so on, because ultimately that might be the only solution. I mean, we we can invent technologies and recommend how they are being used, but then within a corporation they may be uh used in a totally different way because it's the way that they can make money. So there's no way for us to prevent that from happening unless we also are involved in um sort of outreach and um education of the general public and so on, rather than just talking, say, to fellow academics.

SPEAKER_01 38:55

I also think that you know if uh people have you know more than one choice, right? And I think the role of research is to kind of open up that possibility, showing that that's not just one way where things can go, right? I think that's also the role of the futurists, right? Not just presenting one path to the future, you should show multiple paths, and then I think people I I believe would be able to select the path that they think you know are desirable, right? And and I think you know that's I think our role is to opening up all this possibility and hopefully more people are coming toward the the light side, not the dark side of how technology is used. But of course, this full-width challenge, and again, I think the question of incentive is a complex one because how would you actually, you know, this technology I'm not creating a vacuum, right? It's we consume resources, like we're not talking about the environmental impact of this technology now, right? So I think that's also a big part of the equation as well. Like if the resource is limited, what is the incentive and how do we actually build the things that people want? Why also be able to sustain the effort uh in doing that, right? I think that will require more uh idea and and more innovation. Um, and you know, many times when people come and ask me, like, oh, AI is taking people's jobs, like what's the role of humanity in the future? I think that you know there's so many challenges that the AI cannot help us figure out, right? Like the question of like how to use AI, you should not ask the AI, you should ask human because we are the one using it. You know, it's the human future, so it's depending on the human decision, right? We can use technology to help us with this, but uh and ultimately um we still have many challenges to solve. So it's you know still an open opportunity for more people to get involved, I think.

SPEAKER_00 40:38

Yeah. So quick fire question just to finish up then. What is the big change in the way we think of and reflect on our own sense of self, our own identity with AI? How will it change? Like 10, 20 years from now, what what will have changed about the way we think about the self?

SPEAKER_01 41:00

Wow, that's a deep question. I thought of uh what Cherry Turkle has written about many years ago. Uh she wrote about uh the personality workshop, right? How you know when you go into a video game, you can change your personality and and and so on. And how she thought of it as like kind of a a period where you can actually experiment and maybe you come back to the real world, you bring certain aspects back. I hope that in the future, with this technology, our ability to think about ourselves would be multiply, like the multiplicity rather than singularity. I think Audrey Tang, right, the digital minister of Taiwan, said that you know, when you think about singularity is is near, plurality is here. And I think that is uh a really beautiful way of saying that we should celebrate this diversity rather than thinking that there's one ultimate uh way of being correct or being right. Um so I think that's I think that's my hope that in the future the technology will expand rather than diminish our ability to uh appreciate ourselves. Yeah.

SPEAKER_00 41:59

Great point. Yeah. Last word to you, Patsy.

SPEAKER_02 42:02

Yeah, well, I think it will be a very powerful tool for exploring ourselves and changing ourselves in directions that we want to change. That's really my hope.

SPEAKER_00 42:17

I hope that too, as well. Thank you so much for your time. It's been a real pleasure and a privilege to speak to you both. I love your work, it's just so fantastic. And um, everybody can check out where to go and explore much more of it in the show notes. But thank you both of you for joining me today. Thank you for listening to The Future of You, hosted by me, Tracy Follows. Be sure to check out the show notes for more info about the topics we covered today. If you enjoyed this episode, please like and subscribe wherever you get your podcasts. And if you know someone who would love this episode, please share it with them. For more on the future of identity in a digital world, visit futuremade.group slash the future of you. To explore the future of everything else, head over to futuremade.group. The Future of View podcast is produced by Big Tent Media.