Microsoft Research Podcast

Microsoft Research Podcast

An ongoing series of conversations bringing you right up to the cutting edge of Microsoft Research.

Art + Architecture + AI = Ada with Jenny Sabin and Asta Roseway

October 30, 2019 | By Microsoft blog editor

Jenny Sabin and Asta Roseway on the Microsoft Research podcast

Episode 96, October 30, 2019

Jenny Sabin is an architectural designer, a professor, a studio principal and MSR’s current Artist in Residence. Asta Roseway is a principal research designer, a “fusionist” and the co-founder of the Artist in Residence program at Microsoft Research. The two, along with a stellar multi-disciplinary team, recently completed the installation of Ada, the first interactive architectural pavilion powered by AI, in the heart of the Microsoft Research building in Redmond.

On today’s podcast, Jenny and Asta talk about life at the intersection of art and science; tell us why the Artist in Residence program pushes the boundaries of technology in unexpected ways; and reveal their vision of the future of bio-inspired, human-centered, AI-infused architecture.

Related:


Transcript

Host: We’re back with another two-guest episode of the Microsoft Research Podcast, this time featuring the co-founder of the Artist in Residence program along with MSR’s latest Artist in Residence. Together, they discuss their current collaboration, Ada, and inspire us to think big about the future of human-driven, AI-fueled, cyber-physical architecture. 

Host: You’re listening to the Microsoft Research Podcast, a show that brings you closer to the cutting-edge of technology research and the scientists behind it. I’m your host, Gretchen Huizinga. 

Host: Jenny Sabin is an architectural designer, a professor, a studio principal, and MSR’s current Artist in Residence. Asta Roseway is a principal research designer, a “fusionist” and the co-founder of the Artist in Residence program at Microsoft Research. The two, along with a stellar multi-disciplinary team, recently completed the installation of Ada, the first interactive architectural pavilion powered by AI, in the heart of the Microsoft Research building in Redmond. 

On today’s podcast, Jenny and Asta talk about life at the intersection of art and science; tell us why the Artist in Residence program pushes the boundaries of technology in unexpected ways; and reveal their vision of the future of bio-inspired, human-centered, AI-infused architecture. That and much more on this episode of the Microsoft Research Podcast. 

(music plays) 

Host: I am thrilled to welcome two guests to the podcast booth today to celebrate the launch of Ada, an AIpowered architectural pavilion in the lobby of Microsoft’s Building 99. With me today are Jenny Sabin and Asta Roseway. Welcome, Jenny and AstaAsta and Jenny, to the podcast! 

Jenny Sabin: Thank you. 

Asta Roseway: Thank you. 

Jenny Sabin: Thrilled to be here. 

Host: So let’s kick it off with you, Jenny. I’ve already given kind of a list of what you dobut that really doesn’t unpack it. I’d like to hear more from you… What gets you up in the morning when you’re not creating art here in Building 99? 

Jenny Sabin: What gets me up in the morning, really, is the anticipation of the day. I engage in three integrated roles, sometimes more than that. First and foremost as an educator. As you mentioned, I am a professor of architecture at Cornell University where I teach design studios and seminars in our professional degree programs as well as our advanced degree programs with a focus on emerging technologies and design, and issues of making and digital fabrication and, more specifically, looking at how collaboration across disciplines impacts the way that we think in the architectural design process. Related to that, I run a lab, Sabin Lab, within the College of Art, Architecture and Planning, where I collaborate with a number of scientists, material scientists, biologists, engineers… where we, together, innovate new materials and think about how buildings, perhaps, could behave more like organisms, responding to their environments and to people, most importantly. And we also think about innovating new digital fabrication protocols across multiplelength scales. And I engage a number of students and senior researchers in my lab. Thirdly, I am principal of Jenny Sabin Studio, which, conceptually, is very linked to the topics and ideas that we’re pushing forward in the research in my lab. But it’s the entity where I’m able to take on applied projects to really push scale, to take on commissions and to engage in incredible opportunities like this, participating in the Artist in Residence Program at Microsoft Research. And it has a different rhythm in terms of what we’re doing in comparison to the lab. 

Host: Yeah. 

Jenny Sabin: But that’s what keeps me going, keeps me busy and also gets me up in the morning. 

Host: Asta… first of all, you’re a podcast alum, welcome back! 

Asta Roseway: Hi, again. 

Host: And a fusionist here at Microsoft Research. And for our listeners who don’t know what that means, they can go back to Episode 44 of the Microsoft Research Podcast and hear Asta talk at length about that. But you helped establish the Artist in Residence Program here at Microsoft Research. So, tell us what it is, why it exists and how you go about selecting artists, like Jenny, to collaborate with. 

Asta Roseway: So, the Artist in Residence Program is essentially a platform that enables researchers here at Microsoft Research to collaborate, to brainstorm, to push their technologies forward with the help and aide of a creative, an artist. It sort of helps diversify and push the boundaries of the technologies they work on in unexpected ways. 

Host: Yeah. 

Asta Roseway: And that’s really important, because art is that bridge into our humanity, right? So if we’re building these technologies that have real world impact, it’s essential to have the right kind of brains in the room and for artists to be able to use art as a vehicle to envision possible futures. 

Host: You both situate your work at the intersection of art, science and technology and you both often connect a few other streets and avenues to that intersection like biology and architecture and math… Talk about why you both like this intersection and what interesting things happen here that might not happen elsewhere? Jenny, why don’t you give us a start? 

Jenny Sabin: Well, I think for me it’s several different aspects. One is I’ve always been interested in both math and science, and the arts. And for many years I struggled to find the right glove, so to speak, in terms of bringing what are often seen as disparate fields together. And that’s what led me to architecture. So I’ve always been good at math and enjoyed biology and the sciences, but also, fundamentally, am a maker, you know, I’m interested in how these ideas and technologies are made manifest in the world. Alongside that, in architecture, you know, I was fortunate, early on in my graduate career, when I was studying at the University of Pennsylvania pursuing my M.Arch., to experience the shift from analog production of drawings in the architectural design process to the digital. So, I spent the first half of my education producing drawings with a Mayline, you know, ink on mylar, which for me was a really great transition and kind of segue from my background in the fine arts to architecture. And then the second half, I was writing scripts and learning all of these new processes and studying under really important people, an important mentor of mine, Cecil Balmond, who I eventually taught with. And at the time, we were struggling with how to make sense of these tools and how it was impacting the design process. And so terms like complexity and emergence and dynamical were borrowed by architects to try and make sense of what we were seeing in terms of form. And I became a little bit self-critical in wondering well, maybe we could actually learn some things from our colleagues in the sciences about these terms. And that was a kind of point of departure for some of the very early collaborations and in forming bridges with scientists. 

Host: Was there any tension? I remember the shift from analog to digital in music, and there were the ‘purists’ who said, you know, digital… I won’t say the word they said… but they didn’t like it and they thought it ruined music and the organic nature of it. So was there any of that going on? 

Jenny Sabin: Oh, yeah. Yeah. Big time. But at Penn Design, I think, it was a pretty special place. There was an interesting critical dialogue going on, but there were definitely faculty, and also students, that were not so keen on the shift that was occurring. And also, you know, and early on with my initial collaborations with cell and molecular biologists, there were a lot of naysayers, you know, sort of, what are you doing with these biologists? What… you know, what a waste of time! 

Host: Asta, I want you to go a little deeper, too, on this question of that intersection of all of these seemingly disparate areas. I think there’s a lot more connection than we may have given credit for in the past, but how do you find that playing out in your own work? 

Asta Roseway: So I have a very kind of similar path to Jenny, although everyone has different instantiations of that. But I also grew up in the analog world, fine arts, my parents are artists, using my hands to create, but then I went into graphic design and that became the digital world. Right? And so you live between these two realms and then you come to a place along your path where you’re like, but wait, there are all these other fields that are exploring and using different tool sets to find things. So for instance, in the case of detecting pollution, right? We can build digital devices to do that, but you could also augment chemistry to detect it, too. So I think the appeal of new materials, of integrating that into practice, is the affordance of being able to explore something beyond your standard boundaries 

Host: Right. 

Asta Roseway…right? Really pushing that vision, which is fun, right? This is why you work with new materials because it affords you new shapes, new experiences, new, you know, evolutions of architecture and that’s why we do it here at Microsoft, too. 

Host: Today we’re talking about a large interactive installation. We could call it art. We could call it architecture. We could call it technology… And it’s installed at one of the most famous technology companies on the planet. I’d like each of you to talk about why you think art is an important element in technology and why, in the 21st century, technology is an important element of art. 

Jenny Sabin: Well, I think, fundamentally, art and design provide openings and disruptions and new ways of seeing in the context of technology. In many ways, I like to think about how what I do is not about solving problems, but generating problems that we then, you know, look at. So it’s important as a bridge. It facilitates new perspectives and, ultimately, and with the case of Ada, I think it fosters research and potentially opens up avenues and new questions that otherwise may not be readily at hand. 

Host: Asta, what would you add to that? 

Asta RosewayYou know, essentially, you know, art affords us kind of the new narratives. So sometimes, like, without art, like, you’re just thinking technology for technology’s sake, right? And so when you start to sort of bring in the creative process, you have serendipitous moments and accidents, happy accidents I like to call them, that actually enlighten everyone involved in the project. And so, in the case of Ada, you have multiple individuals who’ve actually risen to the occasion to push beyond their comfort zones, to help instantiate something novel. And that’s fundamentally very, very hard and difficult, but it’s so rewarding on a soul level because you’re pushing out of your comfort zone and it really goes back to that pushing yourself out, but also learning from others. right? Like, we have learned so much from Jenny and her process and how she thinks about the world and how she puts it together. And she, in turn, has also, hopefully, learned things from us in how we process and digest the world. And I think those synergies is what elevates all of us. 

Jenny Sabin: Yeah, no, I’ve learned a ton. 

Asta Roseway: So much. 

Jenny Sabin: Yeah, it’s been an incredible journey. 

Asta RosewayIt really has. It’s like climbing a mountain. 

Host: How? 

Asta Roseway: Well, in that you just one step at a time. And you’re going up in elevation and the higher you get, you look down and you go, oh my god! Like, there’s no turning around! We’re going up, right? 

Host: Free solo. 

Asta Roseway: Yeah, kindaKinda! 

Host: Jenny, there are a number of terms you use to describe your work and the materials that you use. And they include things like adaptive architecture, bio-inspired design, material computation, programmable matter… Asta, you’ve used that too, right 

Asta Roseway: Yeah, I love it! 

Host: …responsive materials. This kind of goes across both of your worlds. They invoke a non-traditional and sort of counterintuitive way of looking at things. Talk about how these concepts inform your approach to academic research, technical innovation and artistic practice. 

Jenny Sabin: You know, in many ways, I think people, when they think about what the architect does, they often times think about the napkin sketch, what we call the parti diagram, that then becomes the big idea for the project that you then design towards. Whereas, in my process, which is where nature is an incredible teacher, I’m interested in sort of turning that upside down and saying, okay, what if that initial sketch is emergent? You know, what if it can change in the context of other relationships, whether those relationships and constraints are coming from materials or us as people, or environment? And so, I start every project without any sort of preconceived idea as to what the final form will be. And that way of designing and thinking has very much been informed by my collaborations with scientists, material scientists, engineers. Because you know in nature, geometry, materiality, issues of context, history, program none of that is separate. It’s all in inextricably linked. And, in the context of what we’re experiencing in architecture, both in terms of big data, but also one of the biggest paradigm shifts to hit us since, you know, some people say, since the Medieval Period, which is that the architects being repositioned as a maker again, right? So we can author scripts and algorithms to communicate in real time with robots, 3D printers, etc. So, our former tools, our notations, drawings, plan, section, elevation and so on, those are still important, but, in a way, they removed the architect from the process as sort of master builder. 

Host: Right.  

Jenny Sabin: And that’s why I’m interested in biology. That’s why I’m interested in looking at these systems. It’s never been about scaling up what might be a beautiful form or cellular form into an architectural form, but to probe what’s behind that image, to look at the processes and the behaviors that have contributed to that morphology. Because I think it’s a really powerful way of thinking in the context of this paradigm shift. And so the terms that you described, they’re terms to define the end products, or the end results, of that design process. 

Host: One of the things that you said – and this actually resonated with me so strongly – was knitting is computational design and one of the first instantiations of 3D printing, as it were. Talk a little bit about that whole concept and what’s already existing that we don’t even think about, but we seem to be wowed by the technological version of it. 

Jenny Sabin: Yeah, yeah. Well and these concepts are the root of the name of the project Ada. I mean, one could say that knitting is one of the first examples of 3D printing, as you say, additively layering link-by-link, row-by-row, material. 

Host: With an algorithm to begin with. 

Jenny Sabin: With an algorithm. I mean some of the 

Asta RosewayI’m never going to look at my sweater the same way! 

Host: I know, right? 

Jenny Sabin: I mean some of the most amazing computational designers that I know are textile designers. You know, they inherently understand a sense of information and how zeros and ones are coordinated into a, you know, a pattern and assembly. And my mom is a weaver. I grew up around looms. So that was something that was a part of the everyday. But there’s an incredibly rich history between the mechanization of the first looms, the invention of the punch card and how that influenced digital space. And so that just blows my mind. You know, to think that something started out as material, a tapestry, a knit, that ultimately influenced digital space and the production of the first computers. 

(music plays) 

Host: Well, let’s talk about Ada. Because she’s our star today. I’m calling her a ‘her and I have permission to do that because it’s an incredibly feminine instantiation of architecture and art and technology… But she’s by far the most ambitious, most visible and most permanent (because I know she gets to live here for two years) Artist in Residence project to date. Because we’re talking on a podcast, and this is a wildly and beautifully visible installation, I’d like you to describe Ada for us, visually and structurally. Who is Ada? What is Ada? And how did she come to life here in the heart of Microsoft’s Building 99? 

Jenny Sabin: Well, technology is fundamentally human and I think, in many ways, the big point of departure for Ada is that. We had a number of really exciting and engaging and inspiring conversations, initially, about artificial intelligence and the possibility of an architectural project bringing form and space to that. And making it playful, making it human, and potentially kind of shifting some of the more dystopic or negative views of AI, the kind of fear of technology, into a paradigm that is really about us and our engagement. And so, the spirit of the project is really celebrating that. And, you know, data has incredible hidden structures and, oftentimes, what may be intangible could, you know, potentially be revealed through our own direct engagement with it. And so the material systems, the form, the way that the responsive materials interact with light that is directly driven, in real time, by sentiment data being collected in the building, that coordinated choreography of data to form to space to our own sort of participation is really what makes Ada a living architecture. And, you know, materiality is something I think is co-produced. It’s not just about the materials responding, but it’s our direct engagement with those materials and the kind of magic that comes forth from that. 

Host: I want you to imagine that I can never get into the lobby of Building 99, or that I can’t see. Describe Ada for me. 

Jenny Sabin: So, Ada is in ellipsoidal form. It is approximately two to two and a half stories tall. It weighs just under two thousand pounds. It’s suspended in an area of the atrium of Building 99. And its material systems entail an exoskeleton composed of close to a thousand bespoke 3D printed nodes that are interconnected with fiberglass rods that shift in diameter depending on the forces that are traveling through it. That exoskeleton is then connected to a fabric structure that’s composed of high-tech responsive fibers. For example, one of the responsive fibers includes photoluminescent thread, which absorbs energy from UV lights. And, once that energy source goes away, it emits that light as knitted light. So there are two surfaces to Ada. The outer surface that connects to the semi-rigid exoskeleton, and a whole series of cells and digitally knit cones that then form a soft surface. And so there are opportunities to peer up through those two surfaces, and you’ll see those connections, the kind of sinewy, knitted cones that, structurally, are acting as springs. So the whole thing works reciprocally, and, ultimately, is suspended so the exoskeleton holds up the fabric structure, the fabric structure holds up the exoskeleton, and then it’s suspended, as I mentioned, at three points within the atrium. 

Host: And what’s inside? 

Jenny Sabin: On the interior of Ada is a large tensegrity cone that entails a sleeve of fiberoptic cables, a bespoke fabric structure, within a whole network of LEDs that run, not only within the tensegrity cone, but also within the network of the fabric structure. So, there are three scales of lighting that are interacting in real time with the sentiment data, the data being collected within the network of cameras throughout the atrium. And that data is then revealed in various zones within the project. Within the tensegrity cone, there’s a single camera on the interior of the pavilion that you can drive directly and will override the other cameras so that you have a sense of actually really engaging with and driving the project in real time. 

Host: Jenny, I want you to go back to these bespoke nodes, because Asta was telling me they’re all different. How many of them are there? 

Jenny Sabin: There’s exactly 895. 

Host: All right, so tell me your process with the math of how this thing was conceptualized and built. 

Jenny Sabin: Yeah. I should mention that there is one axis of symmetry, and so each node has a partner, so to speak, in the project. But I quickly realized that if we were going to make this exoskeleton work reciprocally with the fabric structure, where, you know, every single cell or knitted cone has a home within that exoskeleton, its surface design, they were going to have to be 3D printed. And so this is a big leap in my work from a kind of technical, formal and structural level. There’s a lot of complexity in this project at many, many, many, many levels. There are many parts to this! 

Host: People, you have no idea. 

Asta Roseway: No idea! 

Jenny Sabin: You know, we spent months on just the 3D printed node aspect of the project doing mechanical tests, looking at what types of materials we would need to work with. I mean it was all, you know, really cutting-edge. 

Host: Where did you land with materials on those nodes? 

Jenny Sabin: They are nylon.  

Host: Hard nylon. 

Jenny Sabin: Yeah. Yeah, so structurally they had to be nylon. 

Host: And each one of them has three numbers on it? 

Jenny Sabin: Yeah, so each node has three numbers, which co-relates to the neighbor that it connects to, so they share a number, and then it also relates to the fiberglass rod length that connects one node to the next. So, in terms of the digital generation of the project, Ada would not be possible without the tools that I work with and, you know, the bespoke processes that were able to you know put forward in terms of 3D printing and digital fabrication. And so everything is parameterized. 

Asta RosewaySo one thing that really struck me about this process, especially when I and my team members were helping sort of the installation process, was how physical math became. Like for me, math was always this thing that was a little bit abstract. And I have a little bit of a fear of numbers. So I’m just like, okay But this process actually manifested and brought math into a physical form in a way that gave me so much more appreciation for the perfection and the precision of math and what it enables, in a way that I wish I had had when I was a kid. Because I think I would have gone down a math road because I would have been able to actually pair together the creative vision with the way that math, you know, enables that. And so having those nodes in my hand, with the number sequences? I was like this is physical math. This is actually watching something happen in the physical world, which was amazing. It was really profound for me. 

Host: Asta, I want to go a little deeper on the technical aspects of Ada. And Jenny, you can pitch in here too because I know you guys are collaborating at a deep level, but there’s a lot of cutting-edge research in Ada that comes from people in this building. So deconstruct Ada technically for us and tell us how she works. 

Asta Roseway: Yeah, so Ada exists on a research platform developed by Daniel McDuff, a researcher here, who has been working in sentiment data analysis. And he’s generated algorithms that can capture expression variations from cameras and infer a general sentiment from that. So, like a frown, a raised eyebrow, a smile… and then he kind of puts that to a sort of sentiment state and from there, we translate that data into our PC, our servers, and then that goes into our Raspberry Pi infrastructures, we have about three that drive the LEDs, right? So Jenny and her team have defined the color parameters of those sentiments and then we’ve also created zones of light that actually map to different areas in the building where that sentiment is being captured, right? So for instance in our kitchens. And anyone can opt out and the data is anonymized. So it’s just a general count, right? How many frowns? How many smiles? And then you’d get an aggregate of that and then that kitchen represents as a zone on Ada. And then, as Jenny was inferring, the par lights below it are a kind of a general, a bigger side of those kitchens like maybe half of the building kind of thing. 

Host: Okay. So you could have, like, the angry zone 

Asta Roseway: Oh, yeah. 

Host: …or the happy zone… or the frustrated zone! 

Asta Roseway: Every once in a while you’ll see a streak of red going through it and you’re like, woo, somebody… 

Host: Well, and as we noted, sarcasm is not easy to capture for humans even let alone machines, so it’s probably, I could fake something and make the machine think that I was happier than I was and we’ve also all had that experience walking by someone thinking of something completely different and our facial expression may read something but that it’s not 

Asta Roseway: Yeah. But that’s part of why we have this program, right? Where this is an emerging area, where it’s very early in its instantiations. And this enables us to sort of embed this into a form factor that can actually be interacted with in a way that isn’t just a camera, right? And so it opens up all these possibilities and these narratives like, well, what does it feel like to have an entity responding to you? 

Host: Mmm-hmm. 

Asta Roseway: Responding to everyone around it. And, be it controversial or not, I mean, this is the reason why we need to have art in the conversation because art is fearless like that. And those are the conversations, essentially, that we need to have. 

Host: Right. Jenny, do you have anything to add?  

Jenny Sabin: So, you mentioned feminine that Ada is feminineAnd, you know, I think we’re experiencing a huge shift in architectural design, and the forms that are now possible, with a move from, on a meta level, masculine forms – were used to occupying spaces that have hard edges, Cartesian geometries, borders, boundarie– to one that is much more feminine. And, in many ways, Ada celebrates that. The forms are more organic, they’re softer, they’re open, they’re networks, they’re pliable, they’re engaging and there’s something about that that we as humanresonate with and I’m really obsessed with that space in terms of how it can offer moments of pause and delight and wonder. 

Host: Well, Ada is not just a beautiful work of art, she’s a working research project. What are the core questions you’re asking with Ada, and what do you hope you’re going to learn from the ongoing data collection, analysis and even visualization? 

Jenny Sabin: One of the exciting aspects of the project for me is how it may fundamentally support and inspire and impact fundamental research that’s ongoing here in Building 99 at Microsoft Research. And the architecture of Ada, both in terms of its material systems, its responsivity, but also the architecture of the hardware and software, is purposefully flexible to allow multiple data streams to drive it. And so my hope is that researchers, you know, here get so excited about it that they’re like, I want to test my data with Ada! 

Asta Roseway: And that’s been happening. 

Host: Has it? 

Asta Roseway: Yeah. 

Host: Well talk a little bit about that. 

Asta Roseway: Well, we’ve had a lot of people you know, ask, oh, is there an open API? Can I plug in my data set? Can I explore? And that is the whole point is that it enables our researchers, our engineers, to live in the material world with their data. 

Host: Well, Jenny, we’re accustomed to describing structures that have embedded AI as smart buildings. But another term I’ve heard is something you call living architecture. So, talk about how you perceive the difference here, and what it might mean to the future of architecture and technology even. 

Jenny Sabin: I think that opens a number of questions. I mean one could say that Ada is embedded architecture. You know, I’m really interested in how, as we move forward, materials and buildings potentially can cease to be just things and elements, you know, that we see and don’t really interact with, but ultimately become more like our own sort of cells in our bodies responding to and interacting with affordances in our environments and also with us. You know, I’m very interested in how architecture can be made more personalized. Imagine if you could generate a window on demand. That was fouryear research project in my lab with collaborators in material science. And also, alongside that, what does that mean in terms of large-scale transformations of our building facades in urban environments? 

Host: So, go back to the window on demand… there are a lot of places people work that have no windows. Can you get a window on demand? 

Jenny Sabin: I mean, that’s one of my sort of dreamsThe responsive fibers and the interactivity with light in Ada, it’s building upon years of deep research into topics such as structural color, looking to nature for models where color change, and change from opaque to transparent, happen through the interplay with light and our own human perception. And so when I mentioned, imagine if you could have a window on demand, it comes from a deep inquiry into these particular materials where you could dynamically switch from opaque to transparent. And so the interplay with light, with the photoluminescent fibers, for example, builds upon some of those concepts. So I’m interested, at a fundamental level, how we can design and engineer materials to interact with us and also their local environments, while at the same time, limiting the amount of energy that is input. Like interacting with the sun, interacting with our own engagement with the materials and so on. 

Asta Roseway: When we think about how much energy is wasted to heat our homes in winter, for instance. So, if you have a material substrate that can actually generate its own heat, then you don’t have to have those other dependencies. Like, this is that kind of world, right, that inspires us to also think about how we want to move forward with our own technologies. 

Jenny Sabin: I mean, my hope is that the next iteration of Ada is at the scale of a building façade, perhaps, here on Microsoft’s Campus! 

Host: Ada is a hugely collaborative and multidisciplinary project with many teams involved as we’ve talked about. I want you to talk about the people that were involved in this project. Who brought what to the table and why was it important to have a broad skill set and perspective for Ada? 

Jenny Sabin: Well, I’ll start with my team at Jenny Sabin Studio. You know, first and foremost, we are a ‘we.’ You know, the whole sort of myth around, especially in architecture, that a single person authors a building is sort of ridiculous. But thankfully, we’re moving away from that. The people on my team, Dillon Pranger, he was the project manager sort of overseeing a number of details and really my right hand person when I needed help and assistance, and without him I don’t know what I’d do! John Hilla, he assisted and led a number of iterations on the design modeling, so worked on design and production. William Qian, he was a new intern in my office and probably didn’t quite know that he was going to be dropped into the middle of the Ada installation, which he, absolutely, was so thrilled about. Apparently, I have a reputation amongst my students. They say, you know, Jenny, people say, have you ever been on a Jenny Sabin project? 

Host: Like it’s Survivor! 

Jenny Sabin: You know, because the projects, they always start out with a very sophisticated digital investigation and design process. But there’s also, always, a very analog human component. And Jeremy Bilotti worked on a number of aspects of the software pipeline, working with Daniel and myself, so my team was fundamental to pushing the project forward. In terms of working with Microsoft Research, I mean, it’s just been absolutely amazing. You know, from the beginning conversations with Asta and Mira, their excitement about working with me, and, you know, I think we just felt like we had found kindred spirits. We were like, we have to do something. 

Asta Roseway: Yeah… 

Host: Yeah. 

Jenny Sabin: And early on, you know, the number of different meetings I was able to have with various researchers here, and the openness and excitement about, you know, where we could go and what we could do, and then, eventually working with Daniel. And then, more recently, with the installation, which has been intense Intense! 

Asta Roseway: Climbing that mountain! 

Jenny Sabin: And in all of the, you know, the amazing ways and also the moments where you are like wow, how are we going to get to that next step? But to see everyone in the building and my team just come together, you know, to pull off the impossible has been, you know, incredible. 

Asta Roseway: Fantastic, yeah. 

Host: Asta, talk about the people here…  

Asta Roseway: So, on our side, we have Wende Copfer, who is a creative director for Foundry 99, and also kind of our… 

Jenny Sabin: Amazing. 

Asta Roseway: Amazing! Hulk Smash Wende! And was able to really project manage the complexities around this project, including working with the city of Redmond, working with our building vendors and our engineers, as well as our communications department. Like, all those things, Wende held that together. And then Jonathan Lester is an electrical engineer in the medical devices group and he was our head engineer on all things electric: electrical systems, Raspberry Pis, infrastructure… he was the point person on that. And Daniel McDuff, of course, you know, built the infrastructure that the data that drives Ada, and Mira Lane, who’s a partner in the Ethics group, also, was very pivotal in the beginning of these conversations and how we could get this far. 

Host: And what did you do? 

Asta Roseway: You know what? Like we were talking about before, fusionist? 

Host: Yeah. 

Asta Roseway: I just move between entities and I try to remove obstaclesI try to reverse engineer, communicate between different brain sets. My role is to engage our community to step up for the call and the challenge. And, much like Jenny was saying, like, have you been on a Jenny Sabin project? I get that reputation here. Like, have you worked with Asta Roseway? Because it is truly an adventure! 

Jenny Sabin: In addition to my team at Jenny Sabin Studio, who are absolutely amazing, I’ve been fortunate enough to work with a number of professionals outside of the studio and also at Microsoft Research. So Clayton Binkley, the structural engineer on the project with Arup, without him, you know, it would be very difficult to pull off his expertise in tensile structures. As well, important manufacturers that I have longtime collaborations with: Shima Seiki and Dazian and so on. So to pull a project like this off demands the expertise and commitment of such an incredible diverse set of people.  

(music plays) 

Host: Well it’s time for what keeps you up at night?” here on the Microsoft Research Podcast. And Ada couldn’t have come at a more opportune time given the nature of this installation and other pieces and devices that are endowed with sensing technologies. It raises some big questions and I know that these are at the forefront of everybody’s minds. What are the conversations that Ada sparks, or even amplifies, as we find AI research heading ever faster out of the labs and into our everyday lives? 

Jenny Sabin: Well I think for me, you know, one of the hopes is that it opens up a dialogue and provides new questions around AI that otherwise people might not think of. As I mentioned previously, I think we, early on, were very interested in how the project could shift perspectives around AI from the kind of fearful and the negative to one that is fundamentally human-centered and positive and playful and optimistic. But I think we’ve been very cognizant of the fact that there will be people or visitors, etc. that maybe are not so keen on what Ada is doing and opening up. And that’s very valid. And so those discussions around the concerns, the excitement, and so on, I think are really important and certainly things that we shouldn’t avoid, right? 

Host: Right. Well, what comes to my mind, and I’ll say it out loud, is privacy and anonymity, and what are you going to do with my data?” and am I being surveilled versus interacting? How are you addressing that right now? 

Jenny Sabin: Well, importantly – and this was from the very beginning – the data that’s being collected is anonymous and people can opt out. They don’t need to participate. Or people can opt in. In fact, you can download the software that Daniel has developed onto your own personal PC. So if you wanted to take over Ada for a day, um…! 

Asta Roseway: You could do it from your office! 

Host: So kind of like Instagram feed only AI…! 

Asta Roseway: Yeah, exactly! 

Jenny Sabin: Yeah, uh... 

Host: How do you opt out? 

Asta Roseway: Well so, in the way of the stations that we’ve set up, you can move away from the camera. And we have signs that actually warn people ahead of time like, hey, this is what we’re doing and we’re being completely transparent about it. And if they choose not to, they can just move around the camera. But here’s what I’ll say: there are cameras everywhere. We carry phones in our pockets. And this is something we cannot avoid. So rather than pushing it away and pretending it’s not there, let’s go strong into this and figure out what it’s about, and where the boundaries are, and how to evolve that conversation. And so this is part of why we have the Residence because, as I mentioned before, art has to be fearless. Art is the one that’s going to provoke the conversations that actually need to be had. We’re not hiding anything. Anybody can interface with it, and if they don’t like it, that’s a discussion we’re going to have. 

Jenny Sabin: Yeah, I mean Ada, it’s fundamentally about transparency and, you know, the various platforms and hardware that we all interact with every day, I would argue, are not about transparency… 

Asta Roseway: Not at all… 

Jenny Sabin: …so to speak. And so that, for me, is a key difference, you know, we’re engaging the topic of artificial intelligence and data collection, but bringing a transparency to it that’s about the production of new knowledge and new interfaces and fundamentally bringing it back to us, as people. 

Host: I love the fact that it’s amplifying this conversation… in a pretty way! 

Jenny Sabin: She is beautiful! 

Host: She’s gorgeous! Not many guests on this podcast were raised by artists. And both of you were, which explains a lot, to start with! Tell us the story of how you ‘kindred spirits’ met and how Ada unfolded from that meeting to today. 

Asta Roseway: I had seen Jenny’s work years ago and her work kind of crossed my radar. She had done an installation with Nike that just, it kind of burned a permanent image in my brain. And a few years ago, my colleague, Mira Lane, and I attended a talk at the U Dub, and Jenny was giving a presentation. And, as she was showcasing her work, some of the images that came up were triggering in my mind, like, oh my gosh, this is the artist! So I had an ah-ha moment because one of the things that I have loved so much about Jenny’s work in particular is that it’s a very fine line between what you perceive as being real and not real. It’s a digital-material hybrid. When you look at those photographs and actually you experience her work in real life it is very blurry, those lines. And I thought, what an amazing opportunity for a company like Microsoft that just lives in the cloud, right, lives in the data cloud, to be able to partner with someone like Jenny, to play in those blurry lines between real and not real. And so, Mira and I both ran up to you, and we cornered you, and we said, you don’t know us, but you will! And how would you feel about partnering with a company that would, you know, help instantiate the notion of living architecture? You know, with data. And we are the data company. 

Host: And what did you say, Jenny? 

Jenny Sabin: Oh, enthusiastically… 

Asta Roseway: She said yes, that’s why we’re here! 

Jenny Sabin: …yes! And I’m originally from Seattle and I’m an alumni of the University of Washington. 

Host: Go, Dawgs! 

Jenny Sabin: Which yeah, yeah. And this fundraising event was part of that. And I was invited by a former professor of mine, Jamie Walker, who’s now the head of the Art School at the U Dub. And, you know, so growing up, seeing, you know, the early days of Microsoft and it was just like this really amazing kind of fullcircle return of many aspects of formative years of my life. So, yeah, on many levels, it was a big Yes, with a capital Y! 

Host: As we close, I’d like you to predict the future. Not in the psychic sense, but from a practically predictive perspective. What’s the future of Ada, here at Microsoft, and what might the future of Ada-like installations at the intersection of art, architecture, technology and science look like? 

Asta Roseway: So I’ll start, as far as Ada’s existence here. As I mentioned before, the notion of opening up an API that enables other researchers and teams to explore their data sets and augment Ada is kind of the equivalent of growing up, you know? So, being a child, going into adolescence, going into middle age, it’s a process we’d like to see her have. So, for instance, if we have, you know, an augmented sound experience that we can augment with her over time, you know, it also just opens up new conversations, and I think Ada deserves to be able to evolve. My hope is that Jenny can build more evolved Adas, over time, in our new campus. And that, perhaps, the two entities could talk to each other and have conversations based on what they’re learning and understanding. I think every building is going to want an Ada, to some extent, so I think you’re going to have some building envy, but let’s just see how that goes! 

Host: Jenny, talk to us about how you see things unfolding. I know that you can’t predict the future, but talk about your vision for it. 

Jenny Sabin: Well, first of all, I hope that Ada eventually becomes a building façade. That would be my hope in terms of its next instantiation. Everything, the relationships, the platforms, the materials, all of it could be taken to a more permanent application as a series of elements that composes a building façade. That, for me, would be really, really exciting because, ultimately, I’m interested in how these experimental prototypes and spatial structures and material systems impact our built environments and facilitate more personalized spaces that put the human first in the context of architecture, which isn’t always the case. And so I’m interested in how, you know, in terms of the future, how architecture can operate in a more evolving way. So, as I was describing my design process, where the diagram emerges through relationships that come into the fold, imagine if our buildings could also evolve in that way and, you know, that’s where I go back to nature as a great teacher. Contextual events, environmental conditions, history, program, etc. affects form and I think, no, our buildings won’t be living, growing, literally moving things, but there could be aspects of our buildings that do that. And so the spirit of Ada, I think, is a nod towards that future. But again, really thinking about how that is human-driven and human-centric as a point of departure for design and art. 

Host: Jenny Sabin. Asta Roseway. Thanks for coming on the podcast! 

Jenny Sabin: Thank you. 

Asta Roseway: Thank you for having us.  

Jenny Sabin: Pleasure. 

(music plays) 

Host: To learn more about Ada, and other Artist in Residence projects, visit Microsoft.com/research 

Up Next

Artificial intelligence

MMLSpark: empowering AI for Good with Mark Hamilton

Episode 92, October 2, 2019 - If someone asked you what snow leopards and Vincent Van Gogh have in common, you might think it was the beginning of a joke. It’s not, but if it were, Mark Hamilton, a software engineer in Microsoft’s Cognitive Services group, budding PhD student and frequent Microsoft Research collaborator, would tell you the punchline is machine learning. More specifically, Microsoft Machine Learning for Apache Spark (MMLSpark for short), a powerful yet elastic open source machine learning library that’s finding its way beyond business and into “AI for Good” applications such as the environment and the arts. Today, Mark talks about his love of mathematics and his desire to solve big, crazy, core knowledge sized problems; tells us all about MMLSpark and how it’s being used by organizations like the Snow Leopard Trust and the Metropolitan Museum of Art; and reveals how the persuasive advice of a really smart big sister helped launch an exciting career in AI research and development.

Microsoft blog editor

Photo of Andrew Fitgibbon

Algorithms, Artificial intelligence, Computer vision

All Data AI with Dr. Andrew Fitzgibbon

Episode 80, June 12, 2019 - You may not know who Dr. Andrew Fitzgibbon is, but if you’ve watched a TV show or movie in the last two decades, you’ve probably seen some of his work. An expert in 3D computer vision and graphics, and head of the new All Data AI group at Microsoft Research Cambridge, Dr. Fitzgibbon was instrumental in the development of Boujou, an Emmy Award-winning 3D camera tracker that lets filmmakers place virtual props, like the floating candles in Hogwarts School for Witchcraft and Wizardry, into live-action footage. But that was just his warm-up act. On today’s podcast, Dr. Fitzgibbon tells us what he’s been working on since the Emmys in 2002, including body- and hand-tracking for powerhouse Microsoft technologies like Kinect for Xbox 360 and HoloLens, explains how research on dolphins helped build mathematical models for the human hand, and reminds us, once again, that the “secret sauce” to most innovation is often just good, old-fashioned hard work.

Microsoft blog editor

Artificial intelligence, Ecology and environment

AI for Earth with Dr. Lucas Joppa

Episode 72, April 17, 2019 - We hear a lot these days about “AI for good” and the efforts of many companies to harness the power of artificial intelligence to solve some of our biggest environmental challenges. It’s rare, however, that you find a company willing to bring its environmental bona fides all the way to the C Suite. Well, meet Dr. Lucas Joppa. A former environmental and computer science researcher at MSR who was tapped in 2017 to become the company’s first Chief Environmental Scientist, Dr. Joppa is now the Chief Environmental Officer at Microsoft, another first, and is responsible for managing the company’s overall environmental sustainability efforts from operations to policy to technology. Today, Dr. Joppa shares how his love for nature and the joy of discovery actually helped shape his career path, and tells us all about AI for Earth, a multi-year, multi-million dollar initiative to deploy the full scale of Microsoft’s products, policies and partnerships across four key areas of agriculture, water, biodiversity and climate, and transform the way society monitors, models, and ultimately manages Earth’s natural resources.

Microsoft blog editor