#85 The Impact of AI on UX Design industry
36:04 Share

AI can't really figure out what problems to solve like designers are supposed to do. So AI can't understand the context of the user, understand deep psychology, can't have the empathy. AI doesn't have the humanness to understand this person has this problem and I think we should unpack this problem space and solve this problem for them. So AI is far from doing that. So we're safe in the part about identifying what humans need and how to improve the world for them. ♪

Hello everybody and welcome to the next episode of Honest UX Talks. As usual, my name is Anfisa and I'm here today to talk about AI in design and how to stay relevant with Ioana. And to be very honest, I think Ioana will be mainly talking today. As you might have understood already from our previous episodes, Ioana is working in the company specifically on the AI project and

She's probably become an expert in this area. So I actually wanted to really ask her a lot of questions and also learn something from myself today. And hopefully this conversation will be also useful for those of you who's worried about AI and will it be taking our jobs and what will be happening, how to stay relevant, all those questions. But before doing that, we also wanted to say thank you to our sponsor, Dovetail. And I'll pass the mic to you, Anna, to introduce you something new about it.

Yeah, so I've been diving deeper into Dovetail lately because I'm doing this very heavy research project and it's really helping me explore the capabilities of this tool and essentially make my life easier. And one thing I love about Dovetail, it's integration system. So it totally makes your life easier.

It's super well integrated. I think there's a list of 30 tools or products that you can integrate with. For example, you can get started with your insights by just importing existing findings from all your files, folders, all that one. So like if you have them in Google Drive or in SharePoint, you can directly import them and then start your insights hub from there, like from something that's already existing and work that's been done.

But another amazing thing is that you can discover insights across your data based on where that data sits, right? So let's say you have interviews and recordings. You can transform the video and audio recordings from Zoom, Meet, Teams, whatever you're using into editable transcripts. Or you have feedback from surveys.

You can import high volume feedback, CSVs, connect other survey tools through Zapier and so on. Or let's say you have a usability testing platform that you're using. You can connect Dovetail with the testing tools and bring your customer data in the Insights Hub. Also, there is an integration that helps you with document analysis. So you can uncover themes, topics, research problems by highlighting and tagging text.

across Microsoft Word, PowerPoints, PDFs, and many more. So that's absolutely incredible. In a way, it fetches data from words all over spread, which is a problem we all have, and then putting it under one roof. And then the last part, which I think is probably one of the best, is that once you have this Insights Hub built and now you have clarity, you have things cleaned up, everything is in one place, then what's amazing is that you can easily share all this knowledge and socialize it within your team because...

Dovetail offers integrations with Slack, Confluence, Notion, Trello, Jira. Yeah, just whatever the tool is for managing the design process, you can easily integrate with it and bring all these insights to surface to other people to see and so on. Essentially, it helps you import data, analyze it across different channels, mediums, and share those findings.

And I think that's lifesaver. So if you haven't tried Dovetail yet, make sure to try it. We'll link it in our show notes. It's a super robust tool if you're doing a lot of research. And I'm super excited. I've discovered it because my research process, I have to be honest, was a bit all over the place in the past. And I really had struggles. And now it feels like I'm much more organized and faster, right? Because it gives you a lot of starting shortcuts and starting points to build a robust research plan. So yeah.

Before we dive into the episode, Ioana, how was your last week? Can you tell us about some updates if you have any? Yeah, I'm pretty excited to share that I'm working on an AI product and working on a lot of content about AI. And I do want to announce that I have two upcoming talks. One is this

Thursday in collaboration with Hotjar. It's about how to stay relevant as a UX designer in the age of AI. And it's going to happen from, I think I'm going to link it in the show notes because I don't want to mess up the hours in different time zones. So I have to check. And then the week after I'm going to be doing an Instagram live with Interaction Design Foundation. That's going to be on May 30th where I can answer live questions.

about the world of AI and how it's impacting us as designers. And then I'm also going to be doing a masterclass with Interaction Design Foundation, still on the topic of AI, where I will dive even deeper into very tactical ideas and how to work with AI and how to embed it in your everyday work as a designer and so on. So I have a couple of things coming up, mostly about AI. It's exciting. It's also I'm going to be talking about it today. So I'll be sharing a couple of thoughts around that as well. And yeah, that's what I'm doing lately.

How about you? Oh, that's escalated fast. Like, I mean, we talked that you're collecting information and kind of get into the topic, but now you have so many plants already baking. Yeah, and it's really strange because for the Hot Jar event, for example, in the first 24 hours, 3,500 people signed up.

Which is super intimidating because as you can all imagine, there's no AI expert right now. So I am not an expert in AI for design. I'm just a person that's working with AI, that's designing for AI interactions and reading a lot on this topic, learning from other people who are much smarter than me. So it really feels like my imposter syndrome is through the roof with like 4,000 people joining that event. But hopefully, yeah.

I'm just going to share my own experience, my thoughts, what I've read, what I've learned. So it's going to be from a place of learning together, not that I'm any AI expert, which I won't pretend I am.

So how about you? How was your past week? My past week was actually nothing special. I'm kind of still working on the feedbacks. It's like taking all my time. So I finished the masterclass two weeks ago, but the feedbacks, apparently something I didn't kind of calculate it correctly, taking me longer because I do like the portfolios review, whiteboard and challenge reviews, CV reviews. It helps me a lot because I feel like I'm becoming stronger and stronger on spotting

the things that you should include should include so it's like a very good training for myself as well but at the same time it's a lot of portfolios and a lot of reviews and I would say I spent per one student around five six hours so yeah it's like another training for me kind of so finishing that still kind of not ready at all for the baby to be very honest with you and candid um

We only bought the stroller and my due date is literally in five days. So I have no idea what to expect from that experience. But yeah, stay, I guess, updated. We will see how it goes. Back to the topic. Today we wanted to talk about AI and design and how to stay relevant. My first question to you, Ioana, will be what is going on today in AI?

And why is everybody talking about it? Why is this topic so hot? What's your stake on it? Yeah, it's a great question and a great intro to this conversation for today. I think we're all witnessing and many of us are even part of a revolution in technology. Many people compare it to when the Internet absolutely changed everything. It feels like what's happening today in the world has the potential of changing and it will change for some parts.

I want to make an important disclaimer before I continue with any of my ideas. Nobody knows what's going to happen, like not even the expert experts, the people who are working in AI for 30 years, 40 years, because spoiler alert, AI is not new. We were using AI in our everyday lives last year, two years ago, three years ago. So it's not new. But what is changing now is, let's say, the prominence, the popularity, the curiosity, the

questions, like the entire mindset. And it has to do with all the tools that are popping up every day. Like, of course, GPT had a huge impact on popularizing this topic and making people part scared, part excited for what's going to come. And so it's because the technology advancement is extremely fast right now. Like, I think it's unprecedented how fast new products appear, new technology emerges, like everybody's in this AI race, companies between themselves, it's

states between themselves. There's also the political conversations like US trying to compete with other countries of the world. And so everybody's essentially racing. So the speed of the advancement is what's causing this, let's say, wave of conversations, wave of excitement, because every day we see a new piece of news like AI can now do this and look at what this AI is doing and look at Microsoft launched this and then Google launched that. And so it's just

a lot of things going on and that's why there's a lot of conversation as a reaction to that i think that's exciting and i think it's really healthy to have these conversations and because it's not completely unregulated but i think we're spending much more time putting money into this race like economic interest and not enough money or time and resources in general in the ethics

and laying a healthy foundation, good principles, good guidelines, and fighting the harm that AI could also. There's another conversation. I think we've all been partly exposed to that as well, that people are a bit worried that bad things can happen with AI as well.

Yeah, I think that the conversation can sort of counterbalance the capitalist interest of just moving as fast as possible. And then if the society starts talking and becoming a voice in trying to temper or moderate this advancement, then I think it's healthy and it's a good thing for all of us to have these conversations. I hope you can make something out of my intro.

No, no, it makes total sense. To me, it sounds like we're indeed experiencing some sort of AI hackathon on steroids. Everybody's cooking something, some ideas, some MVP. Exactly, it's insane. Yeah.

And the winner is just some whoever is sticking with the market. And it's true that chat GPT changed a lot. I think with the introduction of chat GPT, we started not to say worried, but understanding what's happening, understanding the opportunities. I myself are using chat GPT for everything that I can come up with.

My husband is writing the master diploma right now. He literally has to generate data from it. Apparently, he has to validate it and double check it and stuff. But you can use it for so many use cases. And it's crazy how much it is already integrated in our lives. And that's just the beginning of it. So yeah, it's insane.

I have two points I want to make about ChatGPT. And so what's funny is that, so how did I end up interested in just very quickly? I'm interested in this space because when I came back from my parental leave at UiPath, the product I was assigned was a product that had an AI component and we're doing essentially semantic understanding of documents.

And so I really had to understand what I'm designing for and understand the space I'm operating in. And so I started spending more time learning and then understanding the challenges that the AI component is bringing to my design process and design decisions and so on. So I spent a lot of time in this mental space.

And also I started playing on the OpenAI playground because we are, we're also using GPT in our product, like most products are doing these days. So essentially I started playing with it before ChatGPT was launched and it felt like I discovered, I don't know, this immense potential, like this huge, amazing secret that's going to just make me a more performant worker. Yeah.

And then like in one month or two months after I started spending time on OpenAI's Playground, ChatGPT was launched and everybody was using it. And I lost my advantage, but that was... Magic power, yeah. I lost my power, like everybody had it now. And this is another important point that I'm going to make throughout this conversation. Everybody will have access to AI. So what will make a difference in the working space

will be how you use it and if you use it. And if you understand that it can be an augment to everything you're doing, to your thinking, to your working and so on. But I'm going to get back to that later. And the second point I wanted to make, to your point about your husband writing his diploma with Chad GPT, turns out that some studies have shown that

People that use ChatGPT for learning are performing better, not because GPT is giving them responses or like shortcuts, but because they can have this conversation with GPT where they end up understanding the concept better. So what's amazing is that now you have your private, let's say, it's far from being a coach because it doesn't have enough agency and it's not necessarily designed for that.

But the mere fact that you can have a conversation on a topic and you can get an interpretation and you can get follow-up explanations and so on, it helps people learn better. So it's incredible because many people are worried, oh, but with ChatGPT, we're not going to be writing our papers. Everybody's going to copy. Everything will be created by a machine and not by our brains. But it turns out that

there's a counter effect to it. Essentially, we're learning better. It really augments our understanding of the concepts we're learning. So that's an interesting point. It reminds me of my childhood, to be honest, because in the childhood, we had class teachers who were saying that you can use cheat lists in our exams, because at least if you're preparing cheat lists, you're actually doing your work and you're studying the class and you're writing it down. So you're remembering it.

And so it is in a way similar concept today that you are, like you said, interpreting information, your data informed to come up with your thesis eventually. So it's pretty interesting. And like, okay, so we just talked a little bit about how chat GPT or AI is changing the educational industry maybe. But I

What about the design industry? How would you say the design industry is impacted by AI at this point? Yeah, again, so I want to like restate it. I'm no expert and I'm not going to pretend I am, but I do spend a lot of mental space learning about it and observing it and using it and so on and talking to people. And so some of the things I'm starting to crystallize or I'm starting to understand are that essentially we're going to see two big shifts.

in the design industry as an effect of the AI revolution. So on one hand, an increasing number of designers will have to design for AI interactions because as we can see, many of the projects we already use like Notion, pick any example, they're adding AI capability.

They're adding AI components to their product. And so AI-based interactions will be more prominent, more prevalent. So essentially, it's just going to create more design problems, more design questions in the good sense for us to solve as designers. And so more designers will end up working and trying to understand the problem space of AI and working in creating solutions that leverage AI and so on. So we all...

Not all of us, but many of us will start designing for AI. So this is one thing that will change. This is a new UI paradigm, if you want. Nielsen Norman just published an article very recently, and they were talking about the fact that this is a completely new UI paradigm that we're witnessing, we're living. And so we all will have to create solutions for this new world, for this new UI, AI-based world. So this is one thing that's going to change, what we work for.

And then the other aspect, which I think it's even more immediate because we can all be a part of it like now, just connect to the internet and use AI, is that we will be using AI in our work. So we will be using AI in the design process. Some of the things can be replaceable.

which is highly debatable, but let's imagine that some of the points in the design process we will be able to replace. Some of the parts of the design process we will be able to enhance, optimize, accelerate, improve, augment, essentially. But other things will still remain very human, which is something that I want to make sure that everybody gets from this episode, is that we will not be replaced by AI, not anytime soon at least. I don't know what's going to happen in 30 years from now.

But in the upcoming years, you're safe. There are some things that you can do to improve your chances of not being replaceable. But essentially, yeah, to close my point, we're going to see two big shifts, which we're already seeing. So it's nothing new. It's just me observing the world. We're designing for AI and we will be designing with AI even more and more and more. That's, let's say, overarching impact that I'm seeing.

It makes sense, like to the point that you said, with the fact that AI might not be replacing us just yet. You know, there was a lot of tools already on the market, even before the Jadubt. And one of the tools I was looking at was this UI Zart. It's been on the market for the last at least few years. And I remember I used it maybe like, I don't know, 2020, and it was not usable to me.

And then I tried and I was like, okay, AI is rapidly changing the landscape. Let's try it again and see if it could stick around in my routine. I'm still a fan of pen and paper. And I sketched out really quickly a few sketches, something very default, nothing too crazy, nothing super sophisticated that will change the world. And I uploaded it to UiZert to see if they could recognize my sketches and come up with like this very quick wireframing. And it was still not there. It was still very, very weak in terms of the output it gave me. And I couldn't use it. I realized that I still have to stick to my current routine.

So far, there is no like one radical tool that completely changed my design routine. Maybe it will. Maybe it will in a few months, maybe one year or something. But as of right now, I still don't see anything hyper revolutionary for the design industry. But like I said, I feel like maybe it will be changing today with this global hackathon happening.

Yeah, that's a very strong point. And I agree. And from all the tools I'm trying and what's going on, tools are not yet there. So like you're saying, you can't really rely. Even ChatGPD, which is super performant and super impressive, what it's generating, sometimes it can, of course, we've all heard about this.

It can hallucinate. Sometimes it gives you an answer that sounds absolutely right because it formulates it in such a way, but it's complete bogus. So the idea is that you still have to check. You still have to do the thinking. And it's the same in the design process. I mean, not one part of the design process can be

fully externalized and trusted and have AI be trusted with it. But that's because it's an infant technology. It's an infant industry as a whole. So it's exactly like in the early days of the internet when you couldn't do much and you couldn't really not even understand how to use it. It didn't seem so helpful.

It had a lot of limitations and so on. So AI is in its early days. And I think that the sooner we start, the sooner we'll be part of understanding how we'll be remembering, right? So you'll be able to tell the story. I remember when I tried my first AI tool and I couldn't use it because it wasn't reliable. It wasn't accurate. And was it trustworthy, right? And now look at where we are. We're probably going to get there. Yeah, I'm sure.

I mean, we're recording it in the middle of 2023. So maybe if you're listening to this episode in 2024, you're feeling like, you guys, you didn't know what you're saying. Anyways, you mentioned already a couple of use cases, but I'm also very interested since we're talking that maybe AI is still in the infant stage, right? And it's not fully ready to replace us or replace even parts of our job. But how do we know what we need to learn, right?

What do we need to do today to stay competitive and relevant in a year from now or five years from now? Are there any directions that we designers can take today? Yeah, so I've been reflecting a lot on that in the past couple of months. And the talks I will be doing with Hotjar and Interaction Design Foundation, they're mostly around career positioning.

for designers in the age of AI. So I could break my answer into parts. One of them is the skills you need to hone, the things that will not be, not in the foreseeable future, replaceable by AI. So that will make you still very valuable and irreplaceable. And then the second part of the answer would be how do you hone them? Tactical pieces of advice on how to thrive in this world as a designer. So essentially what I've been thinking about or just learning from other people who are also thinking about that is

the kind of work that AI can do and the kind of work that AI can't do. And everything is up for debate. That's the thing. So there's no absolute answer in this world because things are changing very fast and hardly anyone can keep up with what's going on. So my first point is that AI can't really think for itself yet, right? Which is also debatable. But...

Bear with me here. It can't really figure out what problems to solve like designers are supposed to do. So AI can't understand the context of the user, understand deep psychology, can't have the empathy, like the humanness. AI doesn't have the humanness to understand this person has this problem and I think we should unpack this problem space and solve this problem for them. So AI is far from doing that.

So we're safe in the part about identifying what humans need and how to improve the world for them. So I could call this, let's say, humanness, figuring out what problems to solve. It has an empathy component, but essentially it's just understanding people. And don't get me wrong, AI knows a lot about people because it reads the internet and then it can generate opinions like people learn like this and people think like that and this is perception and this is whatever. So they have information about psychology.

but they're not able to use that information in any meaningful way. So we are the ones who put the pieces together. And this leads me to my second point, which is that building products is really a multidisciplinary effort. It's like putting several boxes together, several things together, like starting from systems thinking, which is the part that brings everything together. But then you bring together human psychology, creativity, process, testing, observing. And so a lot of things like strategy, UI, analytics,

visual design, perception, and so on. So you put together a lot of puzzle pieces in a system. So this component of systems thinking is something that AI can't yet have, but it's not impossible that it will have. Because right now we're in the age of narrow AI, and we hear a lot of conversations about AGIs. Yeah, right, when it's going to become like sentient almost and think for itself, and AI can do a lot more and make connections and so on, but it's not there yet. So I think

There are conversations about putting different machine learning models together to collaborate, led by an overarching machine learning model that kind of guides this collaboration. So there are things and experiments. And so the industry is heading in that direction, but it's not there yet. And so we are the ones who can put the pieces together. We're still safe in that aspect of systems thinking, critical thinking, if you want. And so critical thinking can't yet be replaced. AIs don't critical think yet.

At least. And then another skill is, that's also very debated, is creativity. And this will be a very philosophical conversation. And I don't think we can time box and really unpack here. Like, what does it mean to be creative? Of course, AI is creative in one sense. It takes things that are already there. It analyzes them and unpacks them and understands how they're built. Like, let's think about images or paintings or whatever. It understands what goes into that. And then it creates something from scratch.

Because all the designs, so even in the intellectual property debate, what's being created is completely new. The learning is based on the work of other artists, but essentially it's a new piece of art. And then it's the same with human artists, right? So we go in art school, we learn how Van Gogh was painting, we learn how Matisse was painting, we learn all these techniques. And then based on what we learn from other artists, we create our own art. So essentially it's comparable in one sense. Essentially AI is

sort of replicating human creativity. It works in a similar way. It learns how to create and then it creates, but that is somehow always misses that human spark. AI creativity will never be as impressive as human creativity because human creativity is based on a person's life, a

a person's experience, a person's suffering, a person's feelings and so on. And I think we're going to see a societal shift in what we value. I know that Adobe Firefly in their product now label the content that's being AI-generated

This is a content by a person. And so you're able to differentiate. And I think that we will all start valuing more human made, just the simple aspect that this was made by a human, right? So an example I love is just the example of Mona Lisa, which is a painting that from a technique point of view can be replicated by AI. AI can definitely generate similar Mona Lisa or a more beautiful, like even a more impressive painting of Mona Lisa.

But what touches us about Mona Lisa is that it's like hundreds of years old. We know the story. The humanity has worshipped it for so many years. It's the humanness aspect of that painting that makes it valuable. And so I think that society will sort of start valuing what's human and not necessarily devaluing what's created by AI, but in a way that will never be able to create the same level of emotion because it doesn't come from suffering. It doesn't come from what makes us human.

Sorry for diverging a bit, but the point I'm trying to make is that the human version of creativity will not be replaceable anytime soon. And I don't think AI can generate quirks and like the interesting solutions really out of the box thinking as much as we will be able to still do. And the last point I'm making about skills is that I'm going to reference in the show notes something that it's called a robot curve that's been proposed by Martin Neumeier.

and he argues that essentially we have four types of work in the world creative work skilled work rote work and robotic work creative work is unique it's a non-routine is autonomous skilled work is like talent driven professional is still very high level let's say it's sophisticated rote work is outsourceable you

you manage it is basically execution and robotic work, absolutely the same. So what you can do to stay relevant in the future based on this proposal is to either move your skills from the base of this model, like from robotic work and road work, move your skills towards skilled work, strategic thinking, creative work, so be more valuable and work towards those areas or design products, work on solving the road work and robotic work problems.

So you either work in that space or you move your skills and become more and more valuable as a professional. And this is, I think, the perfect segue into the second point of my answer, which is what can we do tactically in terms of real piece of advice? What can we do to stay more relevant? I will dive deeper into these topics in the upcoming Hotjar event and in the masterclass I have with Interaction Design Foundation. So if you feel like this is enough or I'm just scratching the surface,

you'll learn a lot more about that in those events. But essentially, I think what we can do to stay relevant and competitive in the market is embrace this revolution, not feel intimidated by it. I think we all have a level of intimidation that stems from a couple of things. One of them is that things are changing very fast. Like even if you try to keep up with what's going on, like new tools are appearing every day. There's this new thing that's going on.

every hour. So it just feels like impossible to race and keep up with what's going on. Like you feel like you don't understand what's going on. I think I feel that and I spent a lot of time in this space. So I think we all feel like we're being left behind when in reality, we're not really being left behind. It's just that it's very accelerated, but the main things will be there to keep up with.

if we want. And then to keep up with all the things, just keep an open mind, be curious, and it can translate into just experimenting with these tools. Use Midjourney, use ChatGPT, use UI Tweezer, maybe even create an experiment every week. Set aside one hour where you just play with these tools and understand how they work and understand their design patterns and understand the principles.

and the interactions and so on and the problems they're solving just create space to experiment and play around this is my best piece of advice like it sounds very straightforward but people don't really make the time to experiment with these tools and I think that's where the learning happens and

And then another thing that you can do is, of course, consume content if you want to learn and understand everything. You can even chat with ChatGPT about how to stay relevant and see what it can help you with. But essentially consume as much content as you can without going overboard because it's noisy already. I can really empathize with opening social media and then AI, AI, AI. Everyone's talking about AI. There are all these experts overnight. I sort of feel bad for content.

contributing to this world of people talking about AI. But I also feel that I'm starting to have some thoughts that are being formulated and I think they're worth sharing.

I can feel how it's noisy and overwhelming. I can compare it to starting in UX design when you're completely overwhelmed by the amount of content that's out there. I'm experiencing the same thing with AI right now. But you could pick a couple of sources and then tune down the other things that feel too loud and then focus on those. So I want to recommend a newsletter. It's called Ben's Bytes. Even

Even that one newsletter can feel overwhelming because sometimes it's like 10 things happened in one day, but it's a good curation of the AI progress. And the last point I want to make is that you can really understand there's a lot of information out there, like with everything and the magic of internet, you can really understand what's going on. Google has, it's called Pear.

And it's their open, let's say, space where they share how they design for AI, some principles, concepts, patterns, and so on. Microsoft has a place where they share principles for designing for AI. So all the big companies have these transparent public conversations about how to design for AI. You could spend some time understanding those. Or you can even take an AI course. There's this course called elementsofai.com.

Elements of AI is free. It's taken by a lot of people and it can really help you understand the basic concepts and basic understanding of what AI is. And then you can also try to figure out where AI is being already embedded in your life because it is. It's in our everyday products we use from Grammarly. It's

been using AI for a long time to suggest text improvements. Any kind of news feed that you're using has an AI optimization of showing you recommended content, and that's an application of AI. I'm not going to even go into self-driving cars and so on, but AI is already in our lives. So what we could do is become more observant of it, like just acknowledge it.

So yeah, those are the things that come to my mind. Well, thank you so much. Of course, we will add resources in the show notes, so you can definitely check it out and also start exploring this industry better for yourself. For me, I totally feel that I'm overwhelmed because I feel like, oh my God, I'm already on the maternity leave and now probably my brain will become like completely...

squirt or whatever happens with the mom's brains and I'll be really you know behind everything that's happening I'm already worried about the fact that I will not be able to catch up it's been hard to catch up with everything that's happening in design industry now it will be impossible for me and I do definitely have those worries as well and so the fact that you just explained and really kind of emphasized where AI in design is really helped me and kind of calmed me down a little bit but you're having a very good point with the fact that we just need to keep

staying curious and trying out different things, setting up one hour aside per week should be able to help you with that overwhelm. And the last question I have for you, Joanna, you mentioned that we have those three robot curve layers, right? Creative work, talent-driven work, and the robot work. Could you maybe think of some specific use cases? Which design work could be easily automated? So which work would be out of job earliest, I guess?

Right. Thanks for asking the question. I think it's very valuable and it's a very important immediate point. So I think what we're already seeing being automated or sort of co-created with AI is the UI part. So obviously, AI can already generate mockups. Maybe they're not perfect. They need a lot of editing. They're not accurate, reliable and so on. But it's happening.

mock-ups, iconography, images, and so on. So the UI part will definitely be, in a way, co-created. Yeah, co-created is the best word. I can't think of a better word. Co-created with AI, generated by AI. And so for these, let's say, very tangible solution spaces, like creating the UI, and I really like this idea.

we will move from being generators of design to being curators of design. So essentially, AI will, let's say, generate 10 concepts, 10 versions of UI, and then we will pick one or two we like, and then we're going to work with AI to refine that concept furthermore, those concepts, and then AI will help us test faster, and then we will have answers faster. So it's a co-creation. AI is a

partner, AI augments our thinking and it helps us like accelerate the things that if we want, again, this is a very important point. If we want that, some people just enjoy moving pixels, making interfaces perfect, starting from scratch. That's great. You don't need to use AI. So you can become the analog version.

of the design process and just do things as you do them today. But if you're tired of like creating components and reusing them, recreating them every time and so on, if you're tired of that, let's say tactical work, then you can absolutely call in AI as a partner and it can do that for you. And you essentially curate and refine along with it.

So this is something that's immediately not replaceable, but it's happening. It's a part of the design process that's being heavily impacted right now. And another part is some people generate ideas, solutions like customer journeys. Some people generate personas with GPT. I'm not sure about that. So I feel that even if you

could like push for using it in this way. I'm not sure it's very reliable because personas need to be based in the interviews you've been doing. And of course, you can do some market research with GPT, right? Like ask them, what do most people prefer buying when they're online or whatever? And it could give you some accurate information. It could also be a hallucination. So careful with that. But that's not a

a part where I feel it's completely reliable. Like you still need to do the thinking and understand the edge cases, understand the quirks of the problem you're solving. And AI can't do that yet. But in the UI space, I think it's a pretty promising partner right now. And other parts of the design process, like I even have in my talks a complete process done with AI, like from doing the market research with ChatGPT to running interviews with other AI to doing just the entire process, how it would look if it would

be externalized to AI, but we're not there yet. Some of the things will happen sooner than faster, like generating UI, high fidelity UI design from paper sketches. We've already been seeing that for a couple of years. Even if it's not very reliable, it's happening and probably it will advance like very accelerated as well. So yeah, I think UI work will be maybe the first one to be sort of replaced or augmented or co-shared responsibility. But the thinking is the one that will be replaced last.

I love this. And I think that's a key takeaway for today's conversation. Just to recap, I guess, creators of design, not recreators of design. Makes total sense. I do hope that this conversation would help some of you to reset your mindset, to look at this from like a healthy perspective of the fact that it's a helper today. It's something that is happening. We won't skip it. We know we won't cut it away from our routines.

But at least in the very first stages, it will be helping us. And then as we keep progressing, keep evolving this technology, we will have to curate it further on. Like adjust right to every new eco-climate landscape of the work. And this is just natural. That's how everything works in life. We didn't have internet 20 to 30 years ago. Now we cannot live without it. It's just a part of the life.

So thank you so much for this conversation. Make sure to check out all the links in the show notes. And of course, if you found this episode helpful, useful with so many resources that Ioana just shared with us, please rate us. We would really appreciate if you can send us rating on Spotify or Apple Podcasts or whatever platform you're using on. And if you have more questions or more specific requests, please keep them coming.

We're definitely making more episodes based on your problems today in the design industry. All right. Thank you so much for listening and have a good day, everybody. Bye-bye. Thank you so much. Bye.