Neri Oxman
(00:00:00) Whenever we start a new project, it has to have these ingredients of simultaneous complexity. It has to be novel in terms of the synthetic biology, material science, robotics, engineering, all of these elements that are discipline based or rooted must be novel. If you can combine novelty in synthetic biology with a novelty in robotics, with a novelty in material science, with a novelty in computational design, you are bound to create something novel.
Lex Fridman
(00:00:30) The following is a conversation with Neri Oxman, an engineer, scientist, designer, architect, artist, and one of the kindest, most thoughtful and brilliant human beings I’ve ever gotten to know. For a long time, she led the mediated matter group at MIT that did research and built incredible stuff at the intersection of computational design, digital fabrication, material science, and synthetic biology, doing so at all scales from the microscale to the building scale. Now she’s continuing this work at a very new company for now called Oxman, looking to revolutionize how humans design and build products working with nature, not against it.
(00:01:13) On a personal note, let me say that Neri has for a long time been a friend and someone who in my darker moments, has always been there with a note of kindness and support. I am forever grateful to her. She’s a brilliant and a beautiful human being. Oh, and she also brought me a present, War and Peace by Tolstoy and Meditations by Marcus Aurelius. It doesn’t get better than that. This is the Lex Friedman podcast to support it. Please check out our sponsors in the description. And now, dear friends, here’s Neri Oxman. Let’s start with the universe. Do you ever think of the universe as a kind of machine that designs beautiful things at multiple scales?
Biomass vs anthropomass
Neri Oxman
(00:01:56) I do. And I think of nature in that way in general. In the context of design, specifically, I think of nature as everything that isn’t anthropomass, everything that is not produced by humankind, the birds and the rocks and everything in between, fungi, elephants, whales.
Lex Fridman
(00:02:19) Do you think there’s an intricate ways in which there’s a connection between humans and nature?
Neri Oxman
(00:02:24) Yes, and we’re looking for it. I think that let’s say from the beginning of mankind going back 200,000 years, the products that we have designed have separated us from nature. And it’s ironic that the things that we designed and produced as humankind, those are exactly the things that separated us. Before that we were totally and completely connected, and I want to return to that world.
Lex Fridman
(00:02:54) But bring the tools of engineering and computation to it.
Neri Oxman
(00:02:57) Yes. Yes. I absolutely believe that there is so much to nature that we still have not leveraged, and we still have not understood and we still haven’t. And so much of our work is designed, but a lot of it is science is unveiling and finding new truths about the natural world that we were not aware before. Everybody talks about intelligence these days, but I like to think that nature has kind of wisdom that exists beyond intelligence or above intelligence, and it’s that wisdom that we’re trying to tap into through technology. If you think about humans versus nature, at least in the realm, at least in the context of definition of nature, is everything, but anthropomass.
(00:03:49) And I’m using Ron Milo, who is an incredible professor from the Weizmann Institute who came up with this definition of Anthropo mass in 2020 when he identified that 2020 was the crossover year when anthropomass exceeded biomass on the planet. So all of the design goods that we have created and brought into the world now outweigh all of the biomass, including of course, all plastics and wearables, building cities, but also asphalt and concrete, all outweigh the scale of the biomass. And actually that was a moment. You know how in life there are moments that be a handful of moments that get you to course correct. And it was a Zoom conversation with Ron, and that was a moment for me when I realized that that imbalance, now we’ve superseded the biomass on the planet, here do we go from here?
(00:04:50) And you’ve heard the expression more phones than bones and the anthropomass and the anthropocene and the technosphere sort of outweighing the biosphere. But now we are really trying to look at is there a way in which all things technosphere are designed as if they’re part of the biosphere? Meaning if you could today grow instead of build everything and anything, if you could grow an iPhone, if you could grow a car, what would that world look like? Where the touring test for, I call this material ecology approach, but this notion that everything material, everything that you design in the physical universe can be read and written to as or thought of or perceived of as nature grown.
(00:05:46) That’s sort of the touring test for the company or at least that’s how I started. I thought, well grow everything. That’s sort of the slogan. Let’s grow everything. And if we grow everything, is there a world in which driving a car is better for nature than a world in which there are no cars? Is it possible that a world in which you build buildings in cities, that those buildings in cities actually augment and heal nature as opposed to their absence? Is there a world in which we now go back to that kind of synergy between nature and humans where you cannot separate between grown and made? And it doesn’t even matter.
Lex Fridman
(00:06:36) Is there a good term for the intersection between biomass and anthropomass, things that are grown?
Neri Oxman
(00:06:36) Yeah. So in 2005 I called this material ecology. I thought, what if all things materials would be considered part of the ecology and would have a positive impact on the ecology where we work together to help each other? All things nature, all things human. And again, you can say that that wisdom in nature exists in fungi. Many mushroom lovers always contest my thesis here saying, “Well, we have the mushroom network and we have the mother trees and they’re all connected, and why don’t we just simply hack into mushrooms?” Well, first of all, yes, they’re connected, but that network stops when there is a physical gap. That network does not necessarily enable the whales in the Dominican to connect with an olive tree in Israel to connect with a weeping willow in Montana.
(00:07:28) And that’s sort of a world that I’m dreaming about. What does it mean for nature to have access to the cloud? The kind of bandwidth that we’re talking about, sort of think Neuralink for nature. Since the first computer, and you know this by heart probably better than I do, but we’re both MIT lifers. We today have computational power that is one trillion times the power that we had in those times. We have 26.5 trillion times the bandwidth and 11.5 quintillion times the memory, which is incredible. So humankind since the first computer has approached and accessed such incredible bandwidth, and we’re asking, what if nature had that bandwidth? So beyond genes and evolution, if there was a way to augment nature and allow it access to the world of bits, what does nature look like now? And can nature make decisions for herself as opposed to being guided and guarded and abused by humankind?
Lex Fridman
(00:08:45) So nature has this inherent wisdom that you spoke to, but you’re also referring to augmenting that inherent wisdom with something like a large language model.
Neri Oxman
(00:08:56) Exactly.
Lex Fridman
(00:08:56) So compress human knowledge, but also maintain whatever is that intricate wisdom that allows plants, bacteria, fungi to grow incredible things at arbitrary scales, adapting to whatever environment and just surviving and thriving no matter where, no matter how.
Neri Oxman
(00:09:14) Exactly. So I think of it as large molecule models and those large molecule models, of course, large language models are based on Google and search engines and so on and so forth. And we don’t have this data currently. And the part of our mission is to do just that, trying to quantify and understand the language that exists across all kingdoms of life, across all five kingdoms of life. And if we can understand that language, is there a way for us to first make sense of it, find logic in it, and then generate certain computational tools that empower nature to build better crops, to increase the level of biodiversity? In the company we’re constantly asking, what does nature want? What does nature want from a compute view?
Lex Fridman
(00:10:11) If it knew it, what could aid it in whatever the heck it’s wanting to do.
Neri Oxman
(00:10:16) So we keep coming back to this answer of nature wants to increase information, but decrease entropy. So find order, but constantly increase the information scale. And this is true for what our work also tries to do because we’re constantly trying to fight against the dimensional mismatch between things made and things grown. And as designers, we are educated to think in X, Y, and Z and that’s pretty much where architectural education ends and biological education begins.
(00:10:51) So in reducing that dimensional mismatch, we’re missing out on opportunities to create things made as if grown. But in the natural environment, we’re asking, can we provide nature with these extra dimensions? And again, I’m not sure what nature wants, but I’m curious as to what happens when you provide these tools to the natural environments. Obviously with responsibility, obviously with control, obviously with ethics and moral code, but is there a world in which nature can help fix itself using those tools?
Lex Fridman
(00:11:26) And by the way, we’re talking about a company called Oxman.
Neri Oxman
(00:11:30) Yeah. Just a few words about the team.
Lex Fridman
(00:11:33) Yeah. What kind of humans work at a place like this? They’re trying to figure out what nature wants.
Neri Oxman
(00:11:37) I think they’re first like you, they’re humanists first. They come from different disciplines and different disciplinary backgrounds. And just as an example, we have a brilliant designer who is just a mathematical genius and a computer scientist and a mechanical engineer who is trained as a synthetic biologist. And now we’re hiring a microbiologist and a chemist, architects of course, and designers, roboticist. So really it’s arc, two of each.
Lex Fridman
(00:12:13) And always dancing between this line of the artificial, the synthetic, and the real, what’s the term for it? And the natural
Neri Oxman
(00:12:21) Yeah, the built and the grown nature and culture, technology and biology, but we’re constantly seeking to ask how can we build, design and deploy products in three scales? The molecular scale, which I briefly hinted to. And there in the molecular scale we’re really looking to understand whether there’s a universal language to nature and what that language is. And then build a tool that I think and dream of it is the iPhone for nature. If nature had an iPhone, what would that iPhone look like?
Lex Fridman
(00:12:59) Does that mean creating an interface between nature and the computational tools we have?
Neri Oxman
(00:13:07) Exactly. It goes back to that 11.5 quintillion times the bandwidth that humans have now arrived at, and giving that to nature and seeing what happens there can animals actually use this interface to know that they need to run away from fire? Can plants use this interface to increase the rate of photosynthesis in the presence of a smoke cloud? Can they do this quote-unqoute “automatically” without a kind of a top-down brute force policy-based method that’s authored and deployed by humans? And so this work really relates to that interface with the natural world. And then there’s a second area in the company which focuses on growing products. And here we’re focusing on a single product that starts from CO2. It becomes a product. It’s consumed, it’s used, it’s worn by a human, and then it goes back to the soil and it grows an edible fruit plant.
Lex Fridman
(00:14:13) So we’re talking about from CO2 to fruit.
Neri Oxman
(00:14:13) Yeah. It starts from CO2 and it ends with something that you can literally eat. So the world’s first entirely biodegradable, biocompatible, bio renewable product.
Lex Fridman
(00:14:24) That’s grown.
Neri Oxman
(00:14:25) Yes, either using plant matter or using bacteria, but we are really looking at carbon recycling technologies that start with methane or wastewater and end with this wonderful reincarnation of a thing that doesn’t need to end up in a composting site, but can just be thrown into the ground and grow olive and find peace. And there’s a lot of textile based work out there that is focused on one single element in this long chain like, oh, let’s create leather out of mycelium, or let’s create textile out of cellulose, but then it stops there and you get to assembling the shoe or the wearable and you need a little bit of glue, and you need a little bit of this material and a little bit of that material to make it water resistant and then it’s over.
(00:15:16) That’s one thing that we’re trying to solve for is how to create a product that is materially, computationally, robotically, novel, and goes through all of these phases from the creation, from this carbon recycling technology to the product, to literally, how do you think about reinventing an industry that is focused on assembly and putting things together and using humans to do that? Can that happen just using robots and microbes? And that’s it.
Lex Fridman
(00:15:48) And doing it end to end. I would love to see what this factory looks like.
Neri Oxman
(00:15:54) And the factory is great too. I’m very, very excited. In October we’ll share first renditions of some of this work and in February we’ll invite you to the lab.
Computational templates
Lex Fridman
(00:16:05) I’m there. I’ve already applied. I haven’t heard back. I don’t understand. Okay. Just before we get to number three, it’d be amazing to just talk about what it takes with robotic arms or in general, the whole process of how to build a life form stuff you’ve done in the past, maybe stuff you’re doing now, how to use bacteria, this kind of synthetic biology, how to grow stuff by leveraging bacteria? Is there examples from the past and explain?
Neri Oxman
(00:16:31) Yes. And just take a step back over the 10 years, the mediated matter group, which was my group at MIT, has sort of dedicated itself to bio-based design would be a suitcase word, but thinking about that synergy between nature and culture, biology and technology. And we attempted to build a suite of embodiments, let’s say that they ended up in amazing museums and amazing shows, and we wrote patents and papers on them, but they were still N of ones. Again, the challenge, as you say, was to grow them, and we classified them into fibers, cellular solids, biopolymers, pigments.