Latest / Real Estate Meets AI / A Conversation with Jennifer Kolstad from Ford Motor Company
Transcript
- Sandeep: Most people think about how a space looks. My guest today has spent a career proving it's about what a space does to your brain. Jen Colstadt is Ford's global design and brand director. She owns the vision for roughly 300 million square feet of physical environment. 10,000 dealerships, but more importantly, to me, what got me so excited? She's also a doctoral researcher in neurosthetics. Who's taken her work everywhere from the UN to people designing habitats off this planet? Today we're talking about the science of how buildings change, what happens when you test it at Ford scale, and how AI fits into it. Jen, welcome to the show. Jennifer Kolstad: Sandeep, thank you for having me. Sandeep: well, your background that is entirely remarkable before all of what I just said, art history, interior design, then masters of architecture, right? And you worked on the Borch Khalifa. And before any of this, I think I heard in some interview that as a young woman, someone said that you weren't good at math. And now you're completing doctoral research at the intersection of neuroscience and buildings. Wha what's happened? Give us give us the whole journey. How are we here? Jennifer Kolstad: It's definitely been a cumulative story, beginning with I'm not good in math, to maybe I am good in math and you know, kind of finding my way through like a creative does. I mean the path just isn't necessarily evident at the beginning of all of this, but you forge your way through and you find what interests you definitely, but also ways that you can be successful as a creative, which is probably the bigger story here. And I've ended up at Ford Motor Company, and I can tell you, looking back at the beginning, I I wouldn't have seen that path clearly. I don't know anybody else who would have said, you know, I wanna be whatever, fill in the blank, a painter, an illustrator, an architect, an interior designer, or whatever, for a Fortune 10 company. It just it was never a thing when I was young. Sandeep: this role has never existed before. You're the you're the first in it for f for Ford. Jennifer Kolstad: Yeah, that's true. That is true. Sandeep: You totally have charted your path. all right. So with that very, very diverse background that you didn't see coming, and here you are doing so many incredible things, I'd love to get your perspective on where the industry is with AI. And I know in our conversations, you've talked about how you've been thinking about AI, and if one is AI averse. And five is AI forward. Where do you think the architecture space is? Jennifer Kolstad: There's so much to say on this topic. okay. Sandeep: Please say it all. Jennifer Kolstad: So I'll start by saying this. Look, design has been on the precipice of change many times before now. You know, AI is another cliff that we're going to courageously have to jump off of. Where we land, I don't know, but it will be a positive landing, I promise you. we were last on this cliff with parametric design, if you recall. Sandeep: Sure do. Jennifer Kolstad: what, 15 years ago, probably now. and everybody was thinking, I don't know if they were like panicking or excitement or a little bit of all of it, but it was like a my god, computers are gonna do all the design work. you know the computer generated form finding was our future of course that was never the case right it Sandeep: I remember that it was not. Jennifer Kolstad: was never the case and it ended up being a very specific type of aesthetic where we were coding and allowing yes you know computational programming to lead us through and things started to kind of look similar you know like a computational design Had a certain aesthetic to it, which was a parametric outcome. And there was, there was never really any fear wrapped around that in the end. There was just a lot of questions and unknowns. and frankly, the BIM story is like that too. That was another cliff. And, you know, we jumped off that one collectively and it landed in good places. The conclusion of that story was never about people losing their jobs and computers taking over and doing the thinking for us. It was just another prosthetic or tool. And AI is another one of those moments. This is a very similar moment, not to be afraid of, to learn as much as we can, to harness the tool and the technology get it working for us. and I would say take advantage of the moment to understand, you know, where we can place our skills and and better ourselves as designers. And for me, this is critical thinking. Definitely the path I've chosen is something around neuroaesthetics. So that's science and data-led design and research. So infusing science in our work. propelling us toward better outcomes, more thoughtful outcomes. That's just there's an opportunity. So Sandeep: Yeah, I I couldn't agree more. It's interesting. There's so much of what you said that I'm I I definitely remember the time where everything was parametrichy and you know, all these like twisty towers and everyone was doing it and everyone was making these shading devices and this, that, and the other. It all kind of looked similar and we pass from that phase. Jennifer Kolstad: Yeah. Sandeep: But the the other thing that you that you said, which I almost go a step further, is that not only are we gonna land in a good place, I do believe that if we don't embrace it, that's where the risk actually lies. Because AI is happening regardless. That is not going back in the bottle. That toothpaste has been squeezed out. It's not going back in. And I think it's contingent upon us as an industry to move with it so we can use it to harness the beauty and the the greatness that it can offer for us versus it happened to us. I think that's where the risk is. Jennifer Kolstad: I often talk about with my team about best and highest use. What is your best and highest use? Where do we best place our resources, our personal resources? and I wouldn't say necessarily that our best and highest use collectively is in producing technical documents, for example. you know, that probably is a great use case to develop AI. Our best and highest use moving forward. is in answering some really difficult questions. And they are becoming increasingly more challenging or difficult. Like questions around, you know, environmental solutioning or what can our buildings do that is like what I was just saying, sort of more tuned for human experience design. You know, how can we achieve these goals? we can place ourselves in these categories of work much more effectively and frankly ask for better billing, better fees, because historically architects, you know, haven't always been paid commensurately to their work. And so there's there's another opportunity space. Sandeep: Yeah, very much so. I think that it's a it's a great way to kind of reset the decision making framework almost so to say, even in a building. So going back to the original question, where would you rate it on a scale of one to five? Jennifer Kolstad: Like AI in general, like rate of song, like or rate a Sandeep: Yeah, AI for architecture. Yeah, like as an industry, are we AI forward? Are we AI averse? Are we somewhere in the middle? If we are in the middle, where are we at? cause again. Jennifer Kolstad: I'm gonna score us just a little below average. So I think there is curiosity because as an industry we are we are curious about the future and we're naturally optimistic. I think we don't know a lot about it collectively. so let's and listen, let's let's break this down into two categories. So let's just call one design. so this is the creative pursuit of architecture. I think there's some monkeying around with AI right now that's landing on some incredibly unintelligent solution. I would just put that in the in a garbage category. I don't know the AI. Sandeep: Sure, let's let's make the the garbage category is is vast. As much as there is good useful use like, you know, good use cases, there is plenty of slop, to be sure. Jennifer Kolstad: There's some slop. Yes. So for garbage, you know, there's some monkeying around and producing a lot of garbage right now. is there any real risk that, you know, to the uneducated that this will, I don't know, take the place of real design thinking? No. No, absolutely not. you know, this is like the HGTV of solutioning right now? We'll we'll put it there. So back in the garbage category, okay. Is there a use case, an argument to be made for the technical side of our work? So producing highly polished technical solutions? Absolutely, yes. Yep. And this is this was where I'll score us a little higher. I do think. Mm-hmm. Yeah. Sandeep: That's the whole construction drawing, is it not? That is all basically like specs, everything. That's a tech that's a technical solution. Jennifer Kolstad: Yeah, yeah. And I think we can collectively rally around ideas like efficiency. you know, because this one draws a straight line to profitability, how we run our businesses as design professionals. This one we get. So there's a strong use case here for us as an industry. Sandeep: Yeah. Yeah. So maybe maybe maybe two then right below average. So we're we're getting up there. Cause I will say that most, you know, I don't know, industry wide surveys are definitely like architecture is probably the the the bottom the bottom ten on AI adoption of all the industries, which to me is not surprising. and I I do I do think that there is some parsing that's necessary on Where can it be useful? So hey industry, let's all let's all rally towards it, you know? it's not all bad and it's not all great. there's some good things. let's start there. Jennifer Kolstad: I mean, one of our challenges is that design is highly subjective. you know, and it's meant to solve for some extremely qualitative questions, not quantitative questions, but the qualitative attributes of environmental conditioning. So we're meant to solve for for that. which means If we just talk about AI for a second or the machine and and when we say AI, let's just take off the table solutions. Really let's focus on building the engine that is meant to solve the problem. Cause that's that's what AI is meant to do. like over time. I talked about the cumulative result of my career. So the cumulative information that becomes the engine that produces an AI something. That's where the hard work is going to be. And each of us brings to the table a proprietary point of view. That's, I mean, that's why you hire us, we designers, we architects. Conceivably, Sandeep: Yeah. Jennifer Kolstad: it would take a lot of time to build that engine correctly. You know, all the Sandeep: Very much so. Jennifer Kolstad: the queries, the information, the experience, the data that produces a something that will start to push us toward a qualitative output. Sandeep: Very much so. And I mean, at least what I have seen is that not only does it take a lot of time and effort and research to produce the brain, the AI brain, to create something meaningful, it's also the the tuning. Each time you use it and provide feedback, then it gets better. It is not like any other, you know, of the large language models that we've seen, just kind of, you know, geared towards a particular task. So switching gears a little bit. Jennifer Kolstad: Hm. Sandeep: I want to learn a little bit more about the work itself from your research. Obviously, neuroaesthetics, like I said, most of our audience has never heard that word. So give me a plain language version. Like what is it? Jennifer Kolstad: Okay. Sandeep: What does it let you know about a building that you didn't know before? Why Break it down for us. Jennifer Kolstad: Let's see. Well, aesthetics is essentially the infusion or the conjoining of science and design. And I I'll even remove data from that descriptor because I mean, a data point is an anything point. It's information. Data can be computational. we're producing data points right now as we're speaking. So let's just take that out because that gets a bit confusing. But science, You know, the scientific method is a hypothesis moving through a phase of experimentation, possibly producing something to test a theory and landing on not a solution, but an answer. And an answer can be right or wrong, and you can iterate and then provide another hypothesis that moves you toward maybe a clearer solution. Narrow aesthetics is essentially bringing that methodology and the science behind it into design or into architecture, thereby allowing us to preemptively substantiate the work. And I don't mean to make that sound more complicated than it is, but it's just saying I have a thoughtful idea about how space should perform. Because I have evidence that it has done that in the past, and therefore I can advance an environmental solution. In a way, it de-risks the work. so I sit at Ford Motor Company, I have business leaders all around me. I have people who invest a lot of money in spaces and places, dealers, let's say, and they want to know what it's going to do. And to be direct, they want to know if it's going to make them money from a retail point of view. And so neuroaesthetics allows us to say it it will very likely it will do this. some of those hypotheses. Are qualitative, not necessarily quantitative. So a quantitative hypothesis might be: you know, your revenue will increase X over the next six months, 12 months, five years. A qualitative hypothesis might be the human connections that are formed inside of the environments will start to look like X. Sandeep: Mm-hmm. Jennifer Kolstad: That takes longer to prove, but it's definitely measurable and provable. And the science piece to that equation helps to get us there. And by science, let me just add, the science that we draw on tends to be, you know, people used to refer to this as the soft sciences or the bridging sciences, but these are, sociologists, neurologists, anthropologists, you know, science in those categories that we're now formally folding into our design teams. Sit alongside our architects. Sandeep: I like how you explain that. To break it down even further, I really like that you even pulled in the the dealership example because it's super relatable. Not that everyone on the audience is gonna have a dealership. We've all likely seen one, been to one, understand what that means as a building type. so for example, I have a dealer. I'm starting a new Ford dealer for City of Atlanta. And you're like, great. And I'm like, I would like to maximize my revenue because that's why I'm making this business plan or whatever. And you're saying that with neuroaesthetic, you can be like, maybe as you're doing the floor plan, or maybe as you're thinking about the interior design, or maybe as you're thinking about the placement of different parts within the dealership, organize it in this way because the likelihood of you to be able to, I don't know, maximize your revenue is highest with this configuration because you're connecting qualitative, not quantitative, because it could be anything and any duration, is connected by this building slash architecture connection. Is is that like you're connecting architecture and design to qualitative outpits that people are looking for. Is that kind of the right framing? Or did I not quite get the nuance right? Please tell. Jennifer Kolstad: You started at the right place. So the net net is revenue, let's say. And then we as human scientists can say, well, one of the obstacles to success might be trust built relationships. You know, like automotive retail environments historically have had a bad rap. And so inside of that, we can identify moments of vulnerability or, You know, customers not trusting certain processes. So, how can we repair that trust using the physical environment as a tool? Well, we know that people feel comfortable in certain scenarios or in certain conditions, and we can start to replicate those conditions to elicit an experience or an emotion that is comfort. so see we can start to reconstruct almost like re-engineering the conditions for the relationships that will lead us toward trust and service and the operational efficiencies that point to revenue. Sandeep: Hmm. Okay, that makes sense. Maybe the very simplistic example that we've all seen of this, and you can tell us that that is correct also, is the Capital One Cafe. Because no one wants to go to a bank. Everyone wants to go to a cafe. Or be like, let's do a Capital Jennifer Kolstad: Yeah, it's a great example. Yeah. Sandeep: One Cafe. Jennifer Kolstad: Yeah, Or like how we improved healthcare environments over the last ten, fifteen years using frankly it was always narrow aesthetics. People didn't call it that 'cause we didn't know that's what it was, but we were able to course correct on a lot of problems in healthcare by designing better environments. And actually I'm just staring at you on screen right now. And look, I can point to a couple of things that are like very simple tools. You you're sitting beside I mean, you put a plant beside you. I don't know who put that plant there, but it probably right, it was meant to kind of make the scene feel a little more comfortable and homey. Well, that's biophilia. There's, you know, you're sitting in front of a wood wall, which is an authentic real material that has grain and tactility. That is biophilia. You know, my my brain is registering that that's a a natural material behind you and it's making me recall certain you know memories and experiences that I've had in my life and so I'm starting to feel a certain way. I see color. Well that's color theory. I'm reacting to your red dress and you're sitting on a a blue chair and my brain is you know is starting to operate and register certain things. But all of this collectively in this little scene that I'm looking at right now This is a neuroaesthetic response that I'm having. Sandeep: Yeah. I love that example to take just even this vignette and be like, Well, let's let's break it apart because it's really cause it's not Jennifer Kolstad: Let's deconstruct this. Yeah. Sandeep: it sounds by the name of it, this very high level you know, black box concept, but really it's what we all experience with pretty much any space that we are in, whether it's our home, our work, th wherever we go, we are constantly registering it, whether or not we put the terminology on it or not Jennifer Kolstad: Yeah, so neuroarts or neuroaesthetics is just naming this now. We're just formalizing, you know, the work that has been there all this time. People call it other things, human experience design, or I just talked about biophilia, but this work has been with us for a very long time. It's just that Sandeep: Very much so. Jennifer Kolstad: the industry, both the scientific and the design industry, are starting to come together in a more formal way. Sandeep: It's amazing. I can't wait for you to be done with your PhD thesis coming to us I say in a in about a year or so, right? Jennifer Kolstad: fingers crossed, here's hoping. Sandeep: here's hoping us and all our audience with you. So you've said that, trying to formalize the next frontier of architecture as designing for performance and not designing to an aesthetic outcome. though obviously it still needs to look good. I can't like, for example, throw these plants on the floor messy. Yeah, please. Jennifer Kolstad: Yeah, let me just talk about that. But so even though we say we're designing for, you know, it's basically form versus function, it's that old argument. We're designing Sandeep: Yeah. Jennifer Kolstad: for function, but that doesn't relinquish us the responsibility of designing excellent, beautiful solutions. And in this world that we are in currently, where images are access to beautiful solutions. are on my phone that I'm holding up right now, by the millions, the baseline expectation is beautiful solutions. So this is this is a blessing and a curse. You must produce award winning, publishable, beautiful solutions with absolutely everything. you will fail if you don't. It's just that we must also do these other things too. Sandeep: Yeah. Jennifer Kolstad: And if any of your listeners know the Kano model or want to do a little Google, the Kano model is a kinetic state of customer expectation over time or customer experience over time. The baseline expectation is beautiful design. Sandeep: Very much so. And if I was to try to bridge it back to the the conversation that we kicked it off with and whether or not I don't know you have considered that as part of your research and I'm sure no doubt, but it's that this base great design, that's a given, very much perhaps a human thought and perhaps as we think about how research or scientific knowledge or I forget the exact framing that you used, helps us put it together. Maybe those are ways that AI can in the future start supporting us or humans can leverage exactly which scientific methods they want to use. And then the iteration, I mean, to me that seems right up AI's alley. Jennifer Kolstad: Yes, Andy, I'm so glad you brought that back. Yes. So this is kind of where I was getting at with the with the vehicle. So right now the vehicle has not been built. However, as we start to create the repository of research, of data points, et cetera, the vehicle will be there for all of us to use to propel our work collectively. And then we're really going to get somewhere. You know, that will become a new baseline. which is this science-led design. So the responsibility at the moment, in my opinion, is to start to build the vehicle, this new tool. obviously I'm biased because I've seen the results of this work. I'm literally tracking it and building what we're calling return on experience, which is the data on the outside of the work so that we can say it did do this. But You know, my contribution needs to be helping to build that vehicle so that all designers have access to it, to a new methodology. Sandeep: Yeah. Yeah. I I couldn't agree more. To me that's basically like the the large language model for architecture. Jennifer Kolstad: Exactly right, yes. Sandeep: Yeah, yeah. I mean, I'll say that being on the front end of that, there's a vehicle that we have built that serves our purposes, but it's not expansive enough to have all the science and all the not gonna use the word data because data can be literally anything, but all all the scientific facts. So there's definitely something there. I love that. Jennifer Kolstad: I mean the challenge is that we need a you, you know, you and your firm, because people like me don't know how to build a vehicle. we've got the data for entry. So there will need to be more firms like Cove, to stand up and give us the tools. It's it's a little bit, you know, when you ask me to score the industry I'll bet my score i is also on the low end because it's It's it's a mystery for me. It the whole thing is so new. Yeah. Sandeep: It's too new. But also it's new. No, that's totally fair. So like you mentioned the Ford dealership example or a dealership example. that was in my mind, your first entry into automobile industry and you know, you'd done hospitals before, you'd done towers before, So just even for the audience, because I don't think it's easy to grasp. the scale of, you know, having this hundreds of millions of square foot of space, workplace, retail hospitality, and you've said that by 2027, you'll have touched ninety percent of the Ford's global workplace. So when you Jennifer Kolstad: Yeah. Sandeep: walked in and saw the portfolio for the first time, where do you even begin? Jennifer Kolstad: People always ask me this question. Yeah. First I'll say Sandeep: Cause it's so vast. Jennifer Kolstad: in this case, so contextually, Ford is the project. It's really no different in my mind than the Burj Khalifa or Yes Island or you know, a these large complex projects where you have a ton of people, a ton of experts coming together. It's very similar by scale, but this time Ford is the project. So you sort of shift your thinking. It's not a building. It's a brand, it's a thing. But you start at the start. And you just, you systematically work your way through the design exercise. And when I say you start at the start, you start with language. Are we speaking the same language? what is the paradigm? Like what do we mean when we say, okay, let's get everybody sort of consolidating their language around design, and then establish Core principles, values, ethos. So, what is the collective goal? We start there. Okay, well, what is the methodology? What does the team look like? You start at the start. And then you just start to move through it systematically. You don't take the whole thing on all at once. You do a project, which becomes an example. We call that a pilot. So is there a pilot that we can really get our arms around? That is a tangible demonstration of what we meant when we were saying the thing at the beginning. You know, it's very hard to teach by saying, to teach by doing is so much more effective. You do your first pilot, then you do another project. In our case, the moonshot was the New World Headquarters, which we've just opened. That was over two million square feet, but we had to kind of cut our teeth a couple of times to learn, learn about human-centered design. Well, you know, is that important to Ford Motor Company? What does that yield? And then get Sandeep: What does cut our teeth mean if you could kind of break that down so the audience would understand, like what does that even mean? Jennifer Kolstad: I had a partner at HKS say to me when I joined there, You're gonna have to hang some skin on the walls. And I thought that was disgusting. But okay, I get what you're saying. Okay. I need to go hunting, I guess, and do some things and do some damage. It's the same thing. You gotta get some projects going, going and completed that you can point to and say, Here's the evidence. It's evidence. So, you know, if we do this right or do this a different way, you might get a different result. Here's the evidence. You do it a few times, you start to build consensus. And certainly at an enterprise at the scale of Ford, where there's a lot of stakeholders and people with varied opinions, it's very hard to argue evidence when you can point to something and eliminate all that subjectivity and say, no, it did do the thing. it did yield X result. You build the consensus and now it's easier, to move forward into the next one, and so on and so forth. Eventually you will achieve cultural and behavioral change. And that's really where you're headed. Sandeep: Love that. Love how you broke it down. so it kinda makes this really vast scale and complex problem, which it is no doubt, but the way that you explain it, it seems a little bit more digestible. Jennifer Kolstad: Or like my mummies to say, how do you eat an elephant? What's the answer? One bite at a time is another very gross example. Sandeep: That's right. Okay, great. So and I would say to that I don't eat elephants. Jennifer Kolstad: Neither do I, but I got it. Sandeep: That's right. okay, cool. So we'll switch gears to something a little bit more light. So this is the rapid fire. Don't think, just answer. Jennifer Kolstad: okay. Sandeep: All right, Coffee or tea? Jennifer Kolstad: Coffee, I'm an architect. Coffee, so much of it. Sandeep: Morning person or night owl? Jennifer Kolstad: Neither. Right in the middle. the sweet spot. So much productivity in the middle. Sandeep: Yes, okay, favorite space you've ever designed and one that you would redo. Jennifer Kolstad: I would redo all of them. Also, I had a boss at the beginning of my career that said, if you ever look at your work and go, God damn, I did a good job, you should not continue. And it's true. There's no space I would look at it and go, Yeah, that was great. Good job. Sandeep: yeah, fair enough. Cause we've definitely stopped growing once we are like, Yeah, I did great. This thing I did two years ago, I was spectacular at it. Jennifer Kolstad: Amazing. God, well done. I would never say that. Sandeep: Most overused word in design right now. Jennifer Kolstad: I hate the word juxtaposition. I really hate it. I kind of hate the word innovation. Well Sandeep: Rejects the position. All right. Yeah. We avoided it today up until just now. Jennifer Kolstad: Yeah, like SmartyPants words. I probably said a few and I really apologize to your listeners. But yeah, just say what it is. Be clear. Sandeep: Love it. AI tool that you use most in the day. Jennifer Kolstad: I don't actually. In Sandeep: dear. Jennifer Kolstad: a day? I guess w I guess we need to talk some more. Sandeep: That's a great sidebar. Jennifer Kolstad: Yes, I guess we should step to the side. Sandeep: one thing you wish every building had. Jennifer Kolstad: like obviously more human centered design. I wish buildings were designed for people. Sandeep: More human centric design. Jennifer Kolstad: Yes, yes. Sandeep: Great. Well, last thing, Jen, you're talking to architects, designers, developers, owners, people who make the built world, and a lot of them are quietly anxious about what AI means for their craft. Some are excited, like you said, some are terrified, like you also said, and most are somewhere in the middle. What's one piece of advice that you would give someone in this profession trying to navigate that change? Jennifer Kolstad: this one's easy. So listen, I will say to your audience, our superpower is curiosity. above all else. be curious. Be curious about this. Ask questions, learn more. I mean, listen to me. I know virtually nothing. But I'm excited. And I yeah, I'm Sandeep: That is hardly the case. Jennifer Kolstad: excited and I can't wait to see where the industry goes because of this. Be curious. Sandeep: Curiosity, that's the best one. Well, Jennifer Kalstadt, thank you so much for taking the time to hop on the podcast today. This was certainly an incredible, jam-packed episode. Thank you for your time. Jennifer Kolstad: It is my privilege.