Achieving Seamless Integration Through User Co-Design

If your team's trying to move faster with AI or scale more efficiently without reinventing the wheel, InfoQ Dev Summit Munich 2025 this October 15th and 16th might help. It's a practical event, talks from people in the trenches, sharing what they've actually shipped, fixed, or learned. No big launches or buzzwords, just useful ideas you can take back to your team the next day. Learn more at devsummit.infoq.com.

Hello, everybody. I'm Olympian Pope and I have in front of me Savannah Konoski and Jenna Fittal from IDEO. Well, if you don't know anything about the company, you can join me as well. I just found a couple of weeks ago that they are the ones that actually brought the mouse to the shape that is currently in. And thank you for that. And without any further ado, I'll ask Savannah and Jenna to introduce themselves.

Thank you so much for having us and for having us at QCon last month. We had a great time. My name is Savannah Knofsky and I'm one of the managing directors of IDEO's Emerging Tech Lab. IDEO is a global design and innovation firm, and so we work all around the world to use the process of human-centered design and design thinking to help. businesses figure out what's next and a lot of the work that jenna and i do in the emerging tech lab is exploring the frontiers of technology

and how to make that technology more useful and more human, and a lot of the time more interesting. And we like to explore the future of technology, both in the process of design and the process of how we make things. and also in the outputs of what we create. And my background is kind of general.

software engineer. Towards the beginning of my career, I co-founded a chain of tech schools and based in Nairobi, Kenya, lived there and ran that for a couple of years. And I've been working in the AI space since before it was cool for over a decade and emerging technology at IDEO. for quite some time too. Jenna? I am co-MD with Savannah of our Emerging Tech Lab.

And we have somewhat similar but a little bit divergent backgrounds. So my academic background is in computational geometry and architecture. So I've kind of always sat right square in that intersection between design and engineering. I've spent a lot of time as a technologist and as a software developer for lots of different kinds of applications, often focused on helping explain complex data to a variety of audiences.

And all of my work is really focused on what new technology means in the context of everyday experiences. So thinking about walking into a room or picking up an everyday object and how those experiences can be. changed through new technologies. That sounds really cool and quite a nice symbiosis. So Jenna is identifying new technology and then Savannah is looking, how can we better interact with that as humans?

I shouldn't ask this, but where do I apply? It's really, really cool. So where do I apply? I'm joking almost. The answer is through the website. Yeah, we have a careers page. It's real.

actually is the best way to come through so yeah excellent excellent usually people think about when you say design they are thinking about Well, broadly speaking, how your website looks like or nowadays how your mobile app looks like or anything else. But in the conversation that I previously had with Savannah, she just said, no, it's much more than that.

So maybe let's start by clarifying what does actually design mean in the frame that you already provided? So, I mean, I think at its highest level, design is about... Figuring out what questions to ask to understand. what people want, what can be made, and how value can be created. So we often talk about these three aspects of what makes for good design being desirability.

feasibility, and viability, which sort of map to those considerations. But I think in my personal practice, I really think of design and engineering as fairly unified and all about that problem definition at core. And that expresses itself differently, whether you're talking about, you know, designing a website or a mobile app, which are important things that we do, or if you're thinking about designing, say, a clinical...

workflow that helps patients and providers spend more time with each other. The kind of base set of considerations can be really similar, even across really divergent outputs. The only thing that I would add is that

We say that everything is designed and everything can be designed. And there's just a matter of kind of how much thoughtfulness and intention was put behind how something was designed. So if we think about like... things that we have in our home that just work really well and fit really well into the context of our lives versus you know other things that we have where we're just like okay it works well enough but it's kind of annoying to use and

whatever I'll, you know, kind of put up with it. We really try to be thoughtful about what we're creating and what the context it's going to exist inside of is so that we're really solving for the like true underlying.

human needs behind the problems that we're exploring and trying to solve with what we create. And specifically when we work in emerging technology, we are exploring far out futures a lot of the time. and trying to map what we think will be technologically feasible or what is technologically feasible with what people want and need. And that second part of the statement is...

The big consideration of the design process is we can start with an exploration of this company wants to use more AI or they want to go in the direction of robotics. But if we start just with the technology, then we don't tap into the true. needs behind the problem that's being solved. And then the products don't sell or they aren't as useful. And so that second part of the equation is incredibly important to us.

To add a little bit onto that, the closed loop of that system is critically important, right? So you can sort of ask yourself strategic questions about what people might want or need. But one of my favorite quotes from one of our founders, Bill Mogridge, is the only way to experience an experience is to experience it.

And so when we design, we're not just wondering what people might like or need. We're taking tangible expressions of even future technologies through mixed methods like simulation and design fiction to give. something to actually respond to feel in some way so that we can, with more confidence, understand if that desirability is actually there. Okay, nice.

still very exciting what you guys are doing so my enthusiasm remains with me still so just put it simple you're just designing the whole thought process into the way how people interact or even the flow with interacting with the given technology. and to make it natural because that's what i'm thinking now most of the things that we use daily and we appreciate the way how we interact with them they are just plain simple they are just extensions of humans

And that's pretty much what you're doing. You're ideating and then probably creating some kind of prototype in order to define the interaction. Okay, that's nice. I'm coming heavily from the technology space. I'm very excited about everything that you're doing. And as I was sharing in our previous conversation with Savannah, one thing remained stuck in my head from my university years, actually my master years, when we had a topic that was called...

Ubiquitous computer interaction. And most of the things were like really far-fetched and it was almost 20 years ago. But now it seems that a lot of the things are not as science fiction anymore as they used to be.

Because, for instance, in my current day job, we're using a lot of 3D printing to just create rapid prototypes, either with resin or the filament ones. And it's very nice to just be able to... to look at it but if i'm thinking about technology i'm thinking about dotworks technology radar but you are even more far ahead of that

Do you have a radar like that where you're just thinking about, okay, this is a technology that is worth looking into. That's something that we'll just leave it for one more year to get more bright or something like that. Obviously, everybody is thinking about AI. Some of them are fearful. So the other ones are very excited. And I suppose most of the other people are somewhere in between. They don't really have an opinion about it. So how do you choose?

your battles because i suppose there are so many things happening out there yeah so there's a few ways this happens a really straightforward one is that we're a client services company so we have clients and they have desires And it's our job to figure out what those desires are just the way that it's our job to figure out what sort of end user desires are as well.

But we also have a really vibrant and robust creative community within IDEO. And we definitely use the power of not just being individual designers to spot trends, to see where the excitement is. is amongst our colleagues in the projects and interests that they pursue outside of our direct client work. And we have a bunch of different structures to support this, one of which is a learning group that meets once a week

for, gosh, coming up on four years that's focused on emerging technology. So sometimes we have external expert speakers from our professional and personal networks. Sometimes we have folks within IDEO who are pursuing some interesting technical angle personally. And sometimes we have client project teams come in and share where they are in their process so that we can all kind of learn to get a little bit smarter and hopefully.

Yeah, see around that corner, which is getting more and more crowded all the time and very full of very specific AI startups. So you can't keep track of them all. And only, I think, through the power of using our collective imagination can we get any kind of a handle on. Yeah, we see technology as building blocks on top of each other. So rather than just kind of focusing on one technology at a time, we think about.

in the future what is going to be technologically feasible because of the fact that other technologies exist so like you know we didn't go straight from the stone age to ai And we think that mixing together various forms of emerging technology is really interesting because then you get multiple forms of advanced functionality at the same time. We also work on really varied timelines.

I have projects that I'm working on right now that are looking at two to three years out and projects that I'm working on right now that are looking at 10 years out. And so the type of kind of mental equations that you're doing to think about what might be possible. at different timelines is also different. And sometimes when you are thinking on those more extended timelines, it's actually about inventing the future and thinking about

Where does this organization want to put its energy and its funding and its momentum to build the future that is going to be exciting for them? And so sometimes that's... I guess always that's contextualized in what's happening in industry and what trends are we seeing and where do we think that things are headed. And also sometimes that future, just because it's so far out, it just doesn't exist yet. And we can't say, oh, this.

company is doing this interesting thing or, you know, we saw this interesting thing in an R&D lab in a university. And so we think that that's going to be the future. Sometimes it's really like, huh, if we squint at... what we think might be possible with technology 10 years from now, this company actually is going to go out and invent that thing. And that's a really fun and interesting place for us to be. And so we actually have a whole practice at IDEO that is

called Futuring. And we use the processes of speculative design. So kind of creating concepts that are speculative in nature and futuristic feeling to explore like. both what the technology might look like in the future and then also just what life might look like in the future. And so we're paying attention to all sorts of social, technological, economic.

political vectors of where things are headed and what the world might look like in order to try to almost invent those future worlds. And we obviously take a lot of inspiration from sci-fi because... There have been a lot of amazing authors over the years that have started to create those worlds and set those visions for the tech ecosystem for better or for worse. So that beats the question. Mad Max?

or i don't know star trek for so dystopia or utopia this one was straight for jenna well i mean And personally, I very much hope for Star Trek. You won't be surprised to learn that I've been a Trekkie since childhood. And I think one of the things that I really love about Star Trek isn't so much its utopian-ness, but its focus on... competence. Obviously not uniquely, but kind of unusually amongst very popular narratives. It focuses on a group of professional people doing their jobs mostly well.

I mean, I think like my highest aspiration for what my work does, it's to help in that pursuit. So do I think we'll get to a... post-scarcity socialist utopia soon. Very, very doubtful for many reasons. But I do think that design should reach for encouraging flourishing in some way. And I really like how utopian science fiction like Star Trek explores that.

Well, at the same time, using the tools of dystopian fiction is really interesting, right? Because it pushes you to consider extremes. And I think that is really critical in the design process to not just be... imagining that sort of happy middle path, but also to be imagining what could go spectacularly wrong or spectacularly well and making sure that you have room.

within the systems that you're designing to accommodate that range of outcomes. Okay. Live long and prosper, I suppose. Anyway, I think from my point of view, it would be a good balance in understanding the need to use. our resources carefully. And that's usually, I would put it in a dystopian frame, but also dream big and shiny future as there are the utopian ones. And I think that's the nice balance if that can be thought of.

So then I have to say that I am very jealous on your time budget for your projects, three to 10 years. I think everybody in the software industry only dream about something like that when people are just negotiating for weeks and months. So that's nice. That's how securing the future quite nicely. But moving back, as you said, your company worked a lot. And as you mentioned, I think it started in the 1970s when DeMoss was initially built. Just now featuring the past somehow.

Are there any situations, any technologies that you said this could have done better in this way or some kind of retrospective and learning from the past and the way how to just change the design process that you use? Yeah, we have many iconic products that are things that people have heard of and use. We worked on the Swiffer and the Viewmaster. the Apple Mouse and the Palm V and all of these kind of early iconic products. And that kind of original thinking was taking the process of design.

And then extending that into kind of a methodology for how to do innovation and how to come up with what's next. And so all the time we're coming up with terrible ideas. And the terrible ideas are actually important for us because... We call them sacrificial concepts and we take them into testing with potential users and get their reactions to understand how they feel about them and those terrible ideas.

take us into what the good ideas actually will be. And we definitely have predicted things 10 years in the future. And it's not... Our job to say, you know, here are all of the things and how the world is definitely going to look 10 years in the future. That's just an impossible thing for us to do. But what we do in our future in practice is.

come up with a set of plausible futures, all of which could be possible given circumstances of future technological breakthroughs, potential resources, cultural shifts. those things come true or come somewhat true, then these plausible futures could exist. And so we kind of use multiple plausible futures as those anchor points for

what might be possible in a lot of time for our recommendations. And not all of them are correct. They're really well informed and they're kind of meant to be guardrails and good anchor points. And as... our clients. And as we go through the process of actually trying to build and discover things, it's okay if those shift a bit based on what new technological breakthroughs are happening or, you know, kind of what the resource needs are all around the world.

I'll add one more example because we just got some really exciting news last week, which is that one of our client partners, Teal Health, got FDA approval for the first at-home cervical cancer screening kit. So this is basically a pap smear that you can self-administer. The first time this has ever been approved. And we helped with the design of a few aspects, including...

The kind of unboxing experience and how a person actually comes to learn how to use this new device that they've never seen before because it's never existed. But also actually physically prototyping the mechanism. the physical design of the product itself, which we did during the pandemic when it was quite challenging to get access to prototyping resources. We used a lot of home 3D printing. So you mentioned that earlier.

as something that you're engaged with as well. And I think this is just a really great example of being able to push just a little bit into the future, go a little bit farther to bring something. to market and actually through a clearance process in a really kind of stunningly short amount of time compared to the usual and i'm just so excited to see this on the market and out there helping people i can imagine especially that

This really has an impact given the aggressiveness the cancer is spreading and it's omnipresence. Even me personally, we had a bunch of examples in our family and I know what that means. So kudos for that.

And how do you actually measure impact as a company? Because you're a managing director, so this has to be also on your agenda as well. It's obvious the part with the projects where it's about commercial feasibility, but then we're looking also in impact because you cannot think about the future without the impact. One of my favorite things is when one of our clients gets promoted because of our work. or when new parts of an organization are built because of our work.

A lot of the time it kind of depends on how the work is received by the organization and we try to be really collaborative in the process of design so we're not the types of consultants where you give us a challenge and we go off and we come back. three months later and we're like, here's the thing. We're the type of organization.

that really want to collaborate with our clients to make sure that whatever it is that we are creating and whatever we're recommending is contextualized in the needs of the business. And a lot of the time business leaders bring us in because They're kind of stuck and they don't necessarily have the capabilities to think outside of their current operating model. And a lot of the time businesses are kind of geared up to create.

one thing or a set of things and that's what they've been doing and they do really well but it's hard for them to get outside of that and so We try to work with our clients to have the ability to extend outside of what is currently possible for them, but in a way that still makes sense for their business and makes sense for the types of bets that they want to make.

So one potential key result for you is promotions per year, right? So that's, I'm joking. So since we started the conversation, one thing was going on in my head.

over and over again. As we're talking about future, we're talking about the way how things will evolve, but we are still stuck with the keyboard. So what are we doing here? The next step will be just using our brainwaves to do that. So basically what I'm saying is that for me, it seems that the evolution of technology is getting at a faster rhythm. If we look at 20 years ago, 10 years ago even, things were a little bit slower paced. Now things are happening so fast. How can you embrace?

the challenge that you mentioned where we are doing something appropriately so that we have a proper interaction. Isn't it a challenge for you as a company, as a group? Definitely is, especially because some of that stuff is really highly concentrated in the pre-commercial space, basically.

And so we have longstanding collaborations with some academic institutions, for example, the MIT Media Lab, where we'll often create some speculative work together and we visit each other and sort of learn about. So yeah, some of those brain-computer interfaces, different kinds of haptics, different kinds of XR interfaces. And then we engage in some of that work directly with client organizations as well.

And actually in the realm of new kinds of interactions, we recently wrapped up support of a design sprint, a non-commercial one for the Apple Vision Pro and medical applications. trying to find clinically meaningful interventions that you could use that hardware and the different kinds of interactions that you can create with that hardware, including

I'm going to use a prop for a second. I know this is a podcast, but including taking kind of everyday objects and tracking them and using them as stand-ins for... pieces of medical equipment that you would need to train with. So this was, for example, for inserting a catheter into a large vein. Glass was used as a stand-in for an ultrasound.

And it was really, really exciting to lend some space and expertise to that group to see how younger designers from out in the community, non-IDORs and real clinicians, institutions. like a mass general could come together and understand or imagine new ways of using these technologies for meaningful results, but not necessarily ones that need to be commercially viable yet.

letting us kind of like, again, take that little bit of a leap into the future. Well, that's quite nice. I was just thinking that I think last year I was working with a partner of mine and we are considering doing some kind of consultancy. work with people from the US and then we're just envisioning how nice would it be that we, rather than going back and forth between Europe and the US, we can just use the new VRs.

the Apple Pro in this case, and we'll just kickstart the interaction in person and then probably have a couple of new sessions and then back and forth. And then I was just thinking that it's quite nice to be on the edge. as you are and then just envision different stuff but then when they appear on the market people use them for other stuff as well so it feels a lot like an inception does it ever happen to you

to just look at something that you design and say, no, that's not the way how it was intended. Or it's more loose than that. It's like, okay, you just dream big. You come with an interaction and then you learn from what you see people doing differently. I think that's kind of the point for us, actually, is that we create something and then we watch how people interact with it. And we're like, oh, that's what you actually need. So part of our research that we do.

Oftentimes, and especially when we're designing physical things, IDEO has its roots in the industrial design, although we design everything now from supersonic jets to avocado packaging to... digital government services and beyond, but our roots and especially the practice of design research actually will take us sometimes into people's homes. And we'll bring sacrificial concepts like I talked about earlier. So like early kind of messy concepts that just sort of.

illustrate an idea and those can be you know things that we 3d printed and actual physical objects or you know i've brought in like paper plates as a prop When I was working on a project, there's all sorts of ways that you can kind of just simulate what that end experience would be. And you take those things into people's homes and you're like, oh, actually, this is way too big to fit in your house or your kid immediately broke it.

okay, great, good to know that this is just not going to work. Or you go in and you say, oh, they're actually currently doing this thing in this totally other way that we had never seen before. So for example, we worked on PillPack. which was, I think, later sold to Amazon. But one of the big insights there was, I guess the end product that we ended up with was these like packaged.

daily packs of pills that you take if you have a couple of different prescriptions that you're taking it kind of has them just clustered in there for you One of the big insights that led to that kind of like nice, easy to open plastic packaging was that the team went into an elderly woman's home and they were like, OK, show us your daily routine and how you take your pills. And she walked over to this like tabletop blade.

like like a grinder blade electronic blade and she was like well my hands are too frail to open up a regular pill bottle so whenever i need to open up a new pill bottle i'm like using this like massive blade to grind off the pill cap. And we were like, oh, okay, that makes sense. those pill bottles don't actually work for a lot of people who are taking a lot of pills, which is the older adult population. And so that took us to the form factor of this plastic packaging. So to your question of...

What happens when people aren't using the thing in the way that we expected? We're like, great, that gives us so much information about what the actual need is and how to design it better. And we kind of really look for that information. Okay, that's cool.

Well, because I went in a jeans shop and I asked for talking jeans that I saw them in a presentation and they just shrugged and they just said, okay, dudes, I don't know what pills are you taking, but those don't exist and say, just wait for it. You know, every time I give that talk and I talk about the talking pants. I ask for the phone number of the CEO of Levi's and nobody has given it to me yet. So anybody that's listening, if you have the CEO of Levi's contact number, please reach out.

Yeah, pants can talk. It's totally possible. Yeah, but that was really, really cool. I mean, it was such a nice way. You could have just put on the slide that, well, we have to be careful with what you're doing. We are consuming so much of everything. But then you actually touch on a point that was really cool. Like lately, a lot of the, I don't know what's the proper term for it, probably vintage clothes.

The secondhand clothes are gaining popularity. I'm just thinking about one of the companies in Europe. It became a multi-unicorn just by selling used clothes, and that's something that is quite nice. But that was quite an interesting concept where you're just taking technology, put it in the proper place, you enhance the experience that a buyer has, and then you're just thinking about the future as well.

And I can tell you that my daughter was very excited. She's 10 and she was very excited about the fact and, oh, are those really talking pants? No, but it's just things about the future because at this point she's just... trying to figure out how she can convince her colleagues to make an impact for the planet. And she's just brainstorming ideas. So I'm just thinking that probably you have a huge warehouse where you have all the previous ideas that are stored there just for moving around, right?

Yeah, I sometimes call it a museum in a suitcase. Sometimes we literally put it in a suitcase and take around some of our more provocational ideas. I guess I want to dive into the pants for a second. So what we're talking about here is this sort of speculative piece that we created where we've encoded into the context of a large language model.

back to chatbot, information about a pair of vintage jeans and things about who wore them, how they were created, kind of all of the information that an informed consumer might want to know about before purchasing, as well as some more fun things like what are its opinions about khaki pants?

But part of the motivation for this design wasn't just that large language models had become available, though that was important, wasn't just the social trends of especially younger consumers wanting to shop based on their values. more than brand loyalty, but also some regulations. So especially in the EU, there are important regulations about

the traceability of things like supply chain, labor regulations, and all this other information that now by law needs to be tracked and stored somewhere. And that suddenly created...

a future world where knowing about your pants is not just a theoretical thing, but actually a sort of mandated everyday thing. And so we tried to think about... okay, if that's the case, what new opportunities are there to take that information and make it engaging, usable, and valuable to folks beyond just that sort of regulatory environment?

That, I think, to Savannah's point earlier about how we look for signals in our futuring work, that's really important, right? Because there are some things that will change about the world. And even some of the scarier ones mean that we'll be living in a different world. environment in the future and what sort of secondary consequences do those different worlds have to our daily lives and maybe, maybe one of them is talking pants. We can hope.

Okay, well, I'm sure you can, but you said the EU word and I have to ask you because I'm in the EU, you're on the other side of the pond and... My feeling is that technology companies are not necessarily happy with the EU regulation. And a lot of people are just throwing interesting words towards the EU regulation at points. I did the same, but now...

My optic changed a bit when I see also the benefit. For instance, the AI Act is really focusing more or less on the individual. And how do you feel as a design company? Because somehow... My feeling is that the intention the EU legislation has and your intention as a design company is pretty much the same thing, to just make sure that the end user gets the most value and in the best place. Do you have customers that say,

we have to get to the EU market. That's a pain in the back somehow. Yeah, I think there's a couple of answers to that. We work for companies all around the world. And so they all live in different regulatory environments. And as designers, one of our responsibilities is to understand the constraints of the business. So we're pretty often like learning about new kinds of regulations or standards. based on either industry or location in the world or both.

I think from a design perspective, understanding different regulatory paradigms is like really exciting because you kind of see this like distilled thinking and understanding from at least. representation of a whole society or culture that's been sort of squeezed through their government and regulatory apparatus.

And that's gold. Whether or not we're necessarily like designing to be compliant with the EUAI Act, for example, I think I'm personally really inspired by the framework of the different levels of recognition. regulation based on the impact of the technology, whether it's, you know, for entertainment all the way up through something that might end up creating bioweapons and using that as like a tiered.

understanding of harm reduction like that's very interesting so yeah it's a bit nerdy but i get excited and inspired by different regulations for sure also we have a studio in london and that our studio in London is excited to work on stuff that is in that context and understands it pretty well. Well, we don't talk about London anymore. They decided to leave. So no, I'm joking. Yes.

I think I bumped into one of your colleagues that is in the London office. Great. I don't know if you heard about it, but I'm going to help you with the future and looking at the future.

AI is important, so maybe you would like to look into the AI space and just think about how the people will interact with that. Living joke aside, generative AI opens the box, even though, as Savannah put it, AI is with us for a long time. But now people really understand it. And at this point, we are probably the height of the hype when people are using it for pretty much everything. And, well, my main concern is with ecology of that thing.

a whole different conversation. How do you feel is the current interaction that we have with AI? I know that Savannah had during her presentation had a point on that, that we are at the point where we are between the horse and carriage and the automobile. We don't call it automobile yet. We are just calling, I don't know. autonomous horse or whatever. So any thoughts on this? I think that we're seeing some incredible potential and some incredible capability.

And that is in generative AI and in many other parts of the AI space. And I think that we're kind of in many ways like... at the beginning of this precipice of having more advanced technology as a part of our day-to-day life. I think that we're all used to social media algorithms. you know just having some level of personalization and a lot of the services that we've used so we've gotten the soft launch of these types of services and

There are definitely industries that have been much deeper in this space for quite some time. But from a kind of commercial consumer perspective, I think that it's becoming more and more explicit for us. And I think that as... the devices that we're using and as the services that we're using get better, then people just come to expect those types of experiences, those kind of advanced experiences. And so what is normalized?

happens relatively quickly but I think that in order for us to get there we have to be building and designing truly useful technology and truly useful services that are mapped into what people want and need and also into the context of their lives and their homes and their work or whatever it is that we're creating the device for. whether the interface is voice-based or brain-controlled or just sensory based on how you're feeling or whatever kind of the input method is.

feels like i'm like yes all of that stuff is super exciting but it has to be kind of fit for purpose and fit for context and the technology has to be good because if it's not good then people don't want to use it I think it's more than just don't want to use it because at that point you have a choice.

But what I'm thinking now is that there is a startup in the place where I am and we don't have many that are that innovative. And going back to Star Trek, as we discussed earlier, these guys build a headset, some kind of LiDAR. to help blind people.

And that was really, really nice because that empowers people that usually live in a very narrow space due to the constraints that the society builds willing or unwilling around us. And then it's just... really opening new doors for a lot of people that otherwise will not have this chance and on the other hand i was just thinking about what you mentioned about the social media algorithms

well that's not necessarily a choice now with the rise of llms i think we have a choice where we can actually even though it's a paradox from my point of view have more time and have a better quality because for instance what i'm trying to do now is to automate most of the things that I would normally do by hand. And obviously, I'd be behind my finger saying that I don't have enough time. But now LLMs allow you to do that. But on the other hand, I still have the privacy issue.

If I just drop it on ChatGPT or Cloud or whatever, what happens with my data? And looking at the way how the world evolves these days, it feels that you want to have your data in a safe space if possible. in a suitcase, as you mentioned, the museum in my suitcase. So I think that's also quite important. These days, people are more and more aware. And that pushes me towards the other concern that I was having. A lot of the presentation that you had during your keynotes when I was...

about the gen z generation and the way how they interact with the technology because they were constrained to use it more and probably a lot of them prematurely i'm just thinking about my kids that my daughter just started school and then

Two weeks later, she had to interact with the laptop. It was the first time for her to interact with the laptop to do her studies, even though she didn't know her colleagues. And before, actually a couple of minutes ago, you mentioned older folks. So how does design...

vary from this spectrum. Well, I don't know all the generations. I'm just probably thinking about the boomers and I know the Gen Zs and they know the millennials because I'm part of that generation. But how did the interaction change? It's a super important question because a lot of the time the people who are making the technology are from one or two generations or the people who are like designing the technology are kind of from one or two generations.

The people who are using and benefiting from that technology or being harmed by that technology are sometimes from their generation, but also from others. I think that, you know, there's this thing of different generations liking to like rag on each other. It's like, oh, you know, Gen Z is I feel like every generation is like the generation before is like lazy and whatever, like.

There are all of these just like stereotypes that we make up about other generations and millennials ragging on boomers. And that's a thing. However, if you're designing products. for other generations, or I think we can broaden this out into designing products for communities that are just like different from your own. using the process of co-design is incredibly important and an oversimplified kind of definition of what that is is that

rather than you just designing something for another group of people and then dropping it on them and being like, cool, hope this works. And it's like imbued with all of my values and perspectives on... how I think that this technology should serve your community, using the process of actually designing with the communities that are going to be using or benefiting from the technology that you're creating.

is important and kind of getting input and actually giving folks a seat at the table to come with their ideas and come with their lived experiences and perspectives about how that technology might fit into their lives. One of the first things that we did, and specifically IDEO has a play lab where we do toy invention and we kind of bring the principles of game design and play into.

everyday objects, including toys, but also extending out into other things. One of the first things that they did when Generative AI came out was go to a bunch of Gen Z folks and say, hey, how do you actually want this technology to fit into your lives? You've been so oversaturated by technology through the pandemic. And that was a really key part of your kind of social.

upbringing and the formation of how you're experiencing the world. So we're on the cusp of this other technological revolution and what do you want from it and how do you want it to fit into your lives? And it's going to kind of change your work and your social dynamics. And so how do we build it healthier? ways with you and for you. And the principles that came out of that work were really like, we have had so much technology oversaturating our world that

We're okay with technology. We just want to make sure that it helps us be more human and stay connected to other people and connected to ourselves. And those principles, I think, are actually just incredibly important for any form of designing technology. So I think that there's sort of this like Gen X millennial mentality around tech that it's.

for efficiency and for making things faster and more streamlined. And, you know, I don't want to think just kind of make the decisions for me and take care of it. I think that sometimes is wonderful. It also takes away some of the friction in our day-to-day life that is actually really important for us in ways that we don't necessarily expect and in ways that...

help us stay connected to ourselves and to other people. So I think just making sure that we're getting outside of our own heads and talking to experts and talking to people and working with folks who are going to be potentially using the services that we're creating is the way to go.

Yeah. And I think one thing to add on to that, especially because this is for a more technical audience, is doing that first step of identifying who you actually are designing or building for. Right. Sometimes maybe you're working at a giant tech company and you have billions of users and then that can be tricky. But for most folks in most applications, like you do have an audience or at least you have an aspiration.

of meeting a certain set of, I don't know, behavioral characteristics or place in the world or some other way that you can define hoop in particular you're designing for. And a process I really enjoy is this algorithm bias checklist by the Brookings Institute that you can run before you start a project.

gets pretty serious and granular about who you're designing, who your end users actually are, and what your intended impact on those users might be. And then you can check again after you've done some of your work. And see, like, are you living up to your aspiration? Or have things drifted? And if so, why? And I think using techniques like co-design and having that intentionality when you start can really improve your outcome, not just...

from a market fit perspective, but actually improve the value that you're making for the people who use your product or service in the end. And a follow-up question on this. You mentioned just making sure. that you know exactly whom is the target that you're addressing.

What was going on in my head was at some point I was listening to a design podcast book, I don't remember exactly what, and they were discussing about personas and how faulty can personas be. And they were providing an example. He lives in a castle. He drives a Rolls-Royce in his 60s. And he lives in the UK. And then everybody was like, okay, you can either choose Ozzy Osbourne or at that point Prince Charles, currently King Charles. So that can be really far-fetched. And then I think...

Looking at the way how environments and cities look like these days, more than just generational sampling, I think it's important cultural sampling. How can you manage to get to that particular point of view in such a heterogeneous environment? Yeah, it's a really good question. And I will say that we relatively rarely use traditional personas in our work. We much more often focus on behavior patterns.

Are we trying to serve people who commute by bicycle or like to spend time with their elders or sort of like more action oriented or behavioral characteristics rather than demographic ones? because that's often a more fruitful site to get inspired when you're designing. But you're right, these are not substitutes for understanding. your actual in real life customers and that's why ultimately we basically always talk to individual people while we're designing and even though

You can't talk to every possible customer. You can identify folks who exemplify some of these behavioral characteristics that you're particularly interested in. You can be inspired, you know, directly by Ozzy Osbourne or by... King Charles by talking to them and learning a little bit more about their full lives as people and how that might intersect with your design problem. And yeah, I mean, that's one of the core tenets of human-centered design, I'd say.

Great. So what I hear you saying is don't go after stereotypes like demographics or whatever culture, but look at the habit of the product that you're developing. So go after the tribe rather than...

whatever people living in a given city or something like that. Yeah. I think to an extent we're looking at behaviors and we're looking at actions and there's also a lot of the time, a big difference between what people say versus what they actually do. And so a great use of emerging technology in the process of design is to be able to create kind of simulated environments or to create.

digital experiences that replicate some sort of physical or digital experience that people might be having in a product that you're creating. And then actually, you know, you can talk to them about it and hear about their habits and hear about how the thing might fit into their lives, but then actually prototype that thing and watch what they do.

and see what happens because oftentimes those two things are actually quite different there's something that we talk about which is the say do gap and So we try to lead with bringing prototypes all the time and actually getting people to kind of walk through or move through those prototypes so that we can map their behaviors to what we make.

Let me see if I stole your idea properly and I understand it. So first of all, you go and discuss with them and you ask them nicely about their intent. And then you go on your idea to come back with a bunch of prototypes. And you just come and say, okay.

This is it. Play with it. And then you're observing how they're actually interacting with the prototypes. And that allows you to use... the observations and iterate over what you have and then narrow down to a more perfected prototype let's say do you have something like a stage where you're iterating over the finished product like we have currently in the technology space

a very fast iteration where you have looking like continuous deployment and observability and you're creating that kind of behavior as well. Yes, definitely. And I guess I'll maybe nuance your description a little bit. We try to always bring some kind of a prototype, but not a finished one and not one that's necessarily related to that sort of final outcome, but rather.

prototype that lets us ask a question more deeply so that we can learn something about a preference or a behavior that then informs that final design. But when we're lucky enough, we get to take it. all of the way so i'll tell a story from a few years ago now about an ai product not generative ai though good old-fashioned ai about predicting labor and material needs at fast casual restaurants around the city of Chicago.

And we did some of that exploratory research and prototyping first, but eventually we actually built a machine learning model and we built a deployed front end. and installed it in a bunch of actual locations that were really serving customers around the city and then had continuous check-ins with the frontline workers and the managers at those locations too.

evolved the design until it got to a place where everybody was seeing the kind of information that was appropriate and actionable to them and that our predictions were working well enough to be useful.

We were actually running the model on one of those old Mac Pros that looked like a black cylinder, basically, which was by one of the data scientists' fraud named the Trash Machine. And so people would... wonder like did the trash machine make the new predictions this morning or did it fail for some reason and do we need to restart the trash machine so that one just like really has stuck in my head both as a sort of

fun and exciting moment for us as designers to go all the way through that process, but also just a really compelling and fast process of doing that iteration, making some code changes, deploying the front end, and then immediately. seeing it out in the market within seconds, which is not so unusual in a lot of software deployment practices these days, especially for web apps, but a little bit different for us than our usual iteration cycle.

Okay, nice. And this leads me to probably my final question. Everybody's using Generative AI.

How do you use it? Because it fits for me. What was nice, it was quite easy to build prototypes. It's a matter of... I mean, it's half an hour to just have something, ideating, doing something in practice. Okay. It's not ideal. It needs a lot of polishing, but how do you do it? I know that Savannah had at least one example in her keynote, but...

What are the lessons learned? Does it work and what doesn't? I use it and I see our teams using it in a lot of different ways because we have so many designers who are working across different crafts and methods. But I think the kind of general... message is that we use it to add rigor to our work and to our research and so if there are ways that we can create either more accurate or higher fidelity concepts and prototypes earlier on

or more personalized prototypes earlier on so that we can get more accurate information from our research, then oftentimes we use it in that way. So whether that's like vibe coding a prototype or creating a like... short scrappy experimental video or you know, having a bunch of kind of different concepts that we can generate on the fly and pull in based off of like the preferences of specific people who we might be testing with. All of those are great.

You know, I think that there are the more obvious like day to day things that we'll use it for where we're just kind of asking questions and helping it co-create assets with us, whether that's like written materials or other things. And at the end of the day, I think that it's in a place where it's really useful for some of that kind of scrappier and more experimental stuff and sometimes for more like later stage polished.

stuff but all the time we are coming in with the creative vision and the creative direction and trying to use it more as like a thought partner to us. But our designers and their vision and their agency in the process is the thing that's the most important. And I think if I could make an ask of... Generative AI toolmakers, it is to center kind of the human and the process that people are going through as they're trying to create something and create experiences in the tools that allow them to.

more clearly and more easily articulate that vision of what it is that they're trying to make and to be in the driver's seat of creating those assets so that vision comes to life.

I'll maybe add one more aspect here, which is an aspect of learning. So I'll talk about vibe coding because that's in many ways closest to my heart. We, and certainly, you know, I have, I think, three cursor projects open right now at various levels of completion. And I use these tools as an extension of my existing skill set.

But we have a lot of people who work on digital products at IDEO who don't necessarily have direct experience with building digital products by, you know, putting their hands on a keyboard and writing TypeScript. But we now have this ability and this, honestly, this like in some ways companion who wants you to learn and to know more about processes and expertise that you didn't get.

from your schooling or maybe even from your work experience. And so we have a few folks, including Neil, a person who works in our play lab, who've gone really deep and have built pretty complex pieces of software without that background technical knowledge, but with a ton of curiosity and a real willingness to put in time and hours to understand the problems and to understand really sort of basic. and foundational ideas like debugging or building a backlog and not coming

to understand these things by reading articles, but by actually making them in collaboration with these AI tools. And now when I talk to Neil, for example, I can use some of the shorthand and the basic ideas of software. development and of digital product design in a way that I couldn't have a few weeks ago before he started staying up until 5 a.m. to build and rebuild the Slack bot that he's been working on.

And I don't want everyone to stay up till 5am every night, but I think that there's something to this companionship, this encouragement of learning and exploration that are baked into some of these tools, at least. And I would echo Savannah's sentiment. I hope... Not only that the designers of these tools help to put the users in the driver's seat, but also to have that a nice little, I don't know, gremlin on your shoulder that encourages you to keep going and keep exploring.

Thank you. Well, to be honest, rigor and accuracy would have not been the two of the words that I would associate with Generative A, at least not in this particular point. But yeah, I understand the benefit of, and I'm also using it and my colleagues are using it as well.

And just for our closing, I'll just have to ask you, do you have any plausible scenarios for the technology in the future? I think that was the term that you used, plausible future. So any, not predictions, but directions that we might expect in the upcoming years, maybe? I am excited about more. and humane technology in general. I think that there's the possibility of us having more everyday robotics and kind of just smarter devices and capabilities.

And I think that they could be really cool and really interesting, but I don't think that they need to be these like crazy sci-fi futuristic or like attention grabbing objects. I think that they can just like. be super functional and work really well and look really beautiful and fit nicely into the context of our homes and our lives and be healthy and help us stay more connected to other people. So if I get to put a vision out there, it is.

That. That's nice. It sounds like an Italian village where everybody's connected with our technology. So I will second that. Thank you. Thank you, Savannah. Jenna, how about you? Yeah, I think mine's fairly related. So there's often this really strong barrier between the technical interface. I'm puppeting, basically.

and then the world around me and one thing i'm really excited about at the sort of intersection of ai and robotics and xr is that there's more and more opportunity for our digital interfaces to understand our physical context and for our physical context to influence our digital interfaces. And we see that in lots of different ways already through things like wearables, like watches that have been around for many years.

now through to things like the Snap Spectacles, which are not a commercial product, but a real AR headset that you can wear around outside. And I think that this sort of... blending and this kind of neither privileging the digital experience, not making you stare into your black rectangle all day, but also acknowledging the value that's showing.

otherwise hidden information can bring to your life as a unified experience is really exciting to me. Maybe not totally new, but I think an experience that we've been walking towards for decades and that I think we're going to see a real acceleration in. That sounds like the humans are taking their head out of the rabbit hole, the social media and the internet provided to them and then getting back to the environment around them using the technology in the right space.

That sounds like an evolution from what we currently have to a more closer to what we actually need to improve our lives somehow. Okay. Thank you for your time. It was a real pleasure and I'm looking forward to follow more of your work. thank you thank you so much for having us yes thank you so much