Damian: Hi, Nadia. Thank you so much for joining me today. In this podcast that we record to get people to know the speakers of Hatch Conference. How are you doing today?

Nadia: I'm good. Doing great. A little bit cold here in Amsterdam summer, but I can't really

Damian: We can just pretend we're recording this in October when it's already a little cold, normally not in July.

Nadia: Yes, but I'm good. How are you?

Damian: Pretty good. I'm excited that you're coming. We've been talking for a while and yeah, just wanted to ask a little bit more about, your background and the things that you do. I know that you're founder of AIxDesign a community of people using and reflecting about ai.

It started about three years ago. So definitely before the hype, let's say how did it actually come to be? 

Nadia: Yeah. So quite a bit more actually. Or it's it's hard to say where did it start because it just evolved as a community. But in 2018 I started in this field and so my background is mostly in design, mostly like UX, UI, but also more surface design, strategic design. So design thinking, but not necessarily graphic or visual design. And I was really curious about AI what's the role of AI in design? Like, how is it going to shape design practice and other creative, adjacent like disciplines and the other way around as well, which is like, how does design and how can designers influence sort of the development and trajectory and impact of AI.

So I started with that and I then did my research and ended up releasing the AI meets design toolkit. Which was just like a free, downloadable PDF resource that I just put out on the web. And then a lot of people responded to that, which is then what kind of, was like a starting sign for the community.

Because I realized like there's a lot of people that are interested and want to approach these topics in a different way. 2019 or 2020, I think I opened like just like a slack space for people to meet. And that's been growing out, we started doing some articles and some like online events and I just snowballed into yeah, what we're becoming now.

Damian: nice, very organic community, which is always like the healthiest way of having it, I think. Do you remember when your interest for AI started? How did it come to be? Like even before you started with that community,

Nadia: Yeah. So I think a lot of like people ask me this question and I'm like, I don't know. Exactly. Because, you hear about it and obviously there's also just the depiction of it, even in, in, pop culture, even on like cartoons and stuff

AI or like humanoid like beings.

I know, even as a kid, I was like, that's so sick. But I didn't really, I don't think I really got a grasp of it until I think this was also like 2000. Sweet. 17 or 18 and I was living in Spain at the time in Valencia, and there was a conference there around AI and I somehow ended up there. I can't remember exactly how I ended up at this conference on AI. I think, really opened my eyes to this being a field that also spans all the disciplines rather than just like a computer science.

Damian: Right.

Nadia: And what was really funny was that same moment that this conference was happening, there was the alpha go game playing out like Google's system playing go with I think the world champion and go, and this, also brought up so much discussion because of course there was like AI beating Kasparov in chess, but people were like chess is a very strategic game. It's a very mathematical, procedural kind of game. But like very clear rules and people had always said go is a very intuitive, like very human, like gameplay, that would be really hard to teach the computer.

 I don't know, something shifted, where people were like, Oh God, we have no idea what we're in for. And this machine learning and AI, it's gonna be so much more than maybe what we thought.

Damian: And I think a good approach is also that it's not just for the scientists, right? And not just for the people who have the mathematical and scientific knowledge, but actually. Each one of us in our disciplines, and I think the latest months have been a proof of that. Everyone has to think about what this means.

Nadia: Even more I would say like the most important work in AI right now is not on the technological bit of it, but much more on, yeah, design, implementation, impact ethics and all the other so much that we have to figure out before, but of course there's the arms race and, we're not going to stop that. 

Damian: I remember when I reached to you to do a keynote on the topic of AI and tools, you said I'll do it, but it all moves so fast that I won't finish it until a week before the conference or something like that. That's fine. Is that still the case? But can you give us an idea more or less of what it's going to be about or what you feel that it's going to end up being about?

Nadia: Yeah. There's a few things that I think I'm really interested in now. One is AI in relation to like creativity and how do we like relate to it as creatives and how in a world that is also, entangling like creativity with commodification and livelihood.

Like, how can we, navigate AI to support us and not replace us. But also more inwardly, like looking at the creative process and like how fuzzy ideas are and how, ideas develop. How does AI potentially being part of that process change, the way we think and like incubate thoughts, ideas, concepts.

It's much more about like critically reflecting on what does it mean to yeah, work creatively with systems like this, to let them into your like craft, like how is your, how can you keep that feeling of craft and I don't know, artistic like

When you're working with models like that.

Damian: I guess there's I'd see like a big, two trends of kind of the generative AI and the kind of co pilot AI, what's your weight on where those two are going? Because, as designers, I guess we're. More threatened by the generative one, but more excited about the other one.

But what is your personal take on that?

Nadia: Yeah, exactly. There's now two streams maybe, but I think it's yeah, spectrum of how do we keep agency and autonomy as people and as creatives over these systems. And yeah, what are meaningful ways? So I think we won't move to necessarily like, yeah, so generative AI right now, it's a lot of.

Oh, produce this and then use the outbirds. I'm not like it's computationally like really sick. We can do that creatively. I'm not super interested in that idea. I'm much more interested in this idea of yeah. Co pilot or co creating. So I think that's one is much more interesting. So rather than, it automating things, using the outputs, which is like further commodification of stuff, 

And it's really early in these tools as well. It was like literally July, just a year ago, nobody had I don't know, none of these tools at their disposal. So it's also important to remember like just how quickly this is happening, that actually nobody knows what's going on and like how to deal with this and also experimenting with the interfaces even for these tools, right?

Because the interface is shaped. How we use them, how we interact with them. So I've seen people come up also with like quite experimental ways of prompting. That's not necessarily just put the text and enter, but there's like a visual mix and match collage thing, which then ultimately, prompts the system.

But it's like interacting with it yeah, in different ways, prompting it in different ways, exploring the latent space in different ways, using... The output of these systems, again, it's like input for your process or again, another tool. So I think it's really interesting to look at yeah, really weird, wacky workflows people are coming up with that integrate these tools rather than it's like a one step to this use, use the output.

So I think, yeah, definitely more co piloting model, but I also think there will be. Lots of different sort of ways that we do that, that we haven't seen yet because it's early days.

Damian: Let's say, for the people who have not been up to date or if they felt already they dropped for two months and then suddenly it's too hard to follow. What would you advise? It's a good way to get started without getting lost in all the noise and the possibilities. If you had any recommendations, Yeah.

Nadia: For creators and designers.

Damian: Mhm.

Nadia: Yeah, so I think the best way to learn about stuff and have, figure out what it means and blah, blah, blah is not necessary to read all the headlines and Like god, no, there's way too many of them So I also don't think like it's not necessary and it's not helpful to try and keep up with all the headlines right, like that's just a Lot of information that you have nowhere To put our channel.

So what I would really recommend for someone that's like a bit feeling a bit overwhelmed, which we all are but also, wants to do something. It's just get hands on. And I think the easiest way to do this at this point is to go to Hugging Face demos,

Damian: Mhm.

Nadia: So it's like Hugging Face is this open source community and a platform, and they host a lot of the models that come out.

Across not just image generation, but across all sorts of a I machine learning capabilities. And most of the models have demos. So you can just try it.

Damian: Mhm.

Nadia: A little bit, and you get like a few tries, but I think actually doing it like that and then seeing you get to see how the model performs a little bit, you get to pull those levers and move the parameters to see how does it change the output. So I would say that, and then you can come back and, talk to people about it and exchange ideas and so on.

But I feel like right now a lot of people write and talk and think about it, but they haven't really

Damian: Played with it.

Nadia: had some experience, which is really helpful.

Damian: No, absolutely. And I don't want to get into the whole like, ethics side fully, or our machine's going to take over the world thing because obviously that's been done. But I'm interested in what do you see as a creative, the risks for yourself, for your own, peers what has been discussed in your own community that's beyond the.

The big words of taking over everything and just ruining jobs, etc.

Nadia: Oh, there's so much, and this is my favorite conversation, I think right now there's a lot of like binary rights, either oh, yeah, it's great and we don't have to work anymore, blah, blah, blah, or there's oh, my God, this is the end of everything. It's like neither of those attitudes, I think, are there are very helpful because they disengage you, like you're no longer part of shaping this and you are part of shaping this, right? Like the future is not written. We make it and there's people busy writing it. So it's get, get in to join it as well. I think interesting questions now are we talk a lot that I don't like AI taking away autonomy and agency.

So I'm really interested in working on a project now that's exploring, like, how can we design AI systems to put agency and autonomy of the user or the person using it at the center, which is which we haven't really seen much of because it conflicts with certain interests. of efficiency and profit, right?

But actually, that's a design decision. We could perfectly well design AI systems that are more that operate on very different values. So I think I'm really interested in exploring what is AI, what could it look like, how could it show up in our lives outside of It's very narrow ideology or very narrow scope that we're seeing now, so I think that sort of thinking about how the UX and the interaction design of the systems like shape our behavior and expanding on what that could be something that I think not enough people talk about is who. Who's going to be affected by this? And people are like, aren't you scared? And I'm like I'm not because I think my job is difficult to automate because it doesn't have, a clear process.

But I think it will affect people differently and it will disproportionately affect people that are already at a disadvantage, both in like the bias that shows up in these systems, but also in like automating jobs that now.

Maybe we like outsource but even in the current reality, if it's like people labeling the data like labeling and annotation, like the labor of that, which is very invisible, but AI has a very physical and real, like value chain, like it's not some magic technology that just came from the heavens.

So I think that as well, which is The present reality of it and people's actual lived experience, it's already like it's already having negative impacts and harming people present day. So I'm also yeah, with the community, it's like we're looking to do more work around that. Like, how can we prevent and mitigate like harms?

Occurring at the moment, people's lived experience rather than only, fear mongering the future.

Damian: Yeah, for sure. I think it's super important that these spaces exist, mostly because I feel we are in this kind of like arms race. Where really is release it fast and put it out there and all these companies, which, who should be responsible for things like that are just, they have to be first, they have to, it's really like months or they're gone potentially.

 It's interesting that those spaces now belong to the smaller people and the smaller groups of people who are thinking about this, even if. I imagine that internally it does exist, probably they just are pushed by the market's needs and the necessity to like, be first.

Nadia: Yeah, exactly. There's so many reasons not to do that. I'm sorry. There's so many reasons to just rush, run to the finish line and not ask the big questions like that's incentivized.

So I think it's really important to create spaces where we again, think and talk about AI outside of that.

It's imagine all the world's research around nutrition came from McDonald's.

Damian: Ha.

Nadia: You know what I mean? It's imagine that all the sort of narratives that are in your head around a certain like field. They all come from a very specific angle that has a specific interest and it's like that's where we're at with AI, it's like our whole idea of what it is and what it means is shaped by such a narrow set of ideas.

So I think if anything, there's so much shit to figure out, but if anything, it's that would be a great place to start, right? Because it's like all the things we have to figure out, they touch on different disciplines and different challenges from Labor rights, but also like environmental impact and, like UX, like all across the board.

So it's the first thing we need is for people to realize that AI isn't just this thing that it's now in their head, but it's, we can expand on what that what that means and develop other, like languages around it, narratives around it, approaches to it that, yeah, people in big tech don't have the space and time because they are, in an arms race.

It's like we cannot ask them to do this. We have to do this, those of us who can find and take up those spaces. Yeah, I think that would be a good first step, towards all the other we can't leave ethics in the hands of meta. that's unfair, that's It's like setting yourself up for failure.

Damian: Lastly, I wanted to ask you, besides giving a talk what are you most excited about coming to Hatch? Nice. Absolutely.

Nadia: Oh, good question. I saw, so yeah, like I said, I was already watching the podcast snippets

And I see there's yeah, quite a bit around like design for AI, design of AI and so on. And it's really fun for me to be in a community with other people that are like actually working on this on a, day to day basis.

Yes, so I'm probably gonna pick their minds a lot. And, yeah, just, it's always really exciting to meet yeah, meet other people that are facing the same challenges and exchange ideas with them. So I'm really looking forward to that. And,

Damian: by the time this episode comes out, also probably the. Roundtable signups will be open and yeah, we'll have like small rooms of up to 15 people that can just get together and chat about a topic. And in your case, obviously it would be like AI and related stuff. 

There's nothing like getting in person and getting really like fine with some stuff. Nice. All right. Thank you so much. That was great. And we'll see each other in October.

Nadia: Okay, thank you.

Damian: Thank you.