CEOInterviews.AI
Start App
Get an email when Fei-Fei Li gives a new interview
One email per new interview. Unsubscribe in one click.
Fei-Fei Li
Co-Founder and Chief Executive Officer; incoming Executive Vice President and Chief Scientist, AMD, World Labs

Emerson Collective Demo Day 2026 | Full Program

📅 Nov 16, 2026 Emerson Collective 125 MIN 57 VIEWS 147 SEGMENTS · 18 SPEAKERS
For 10 years, Emerson Collective’s Demo Day has brought together bold thinkers, creators, and innovators from across the EC community to share the work they’re doing to tackle some of the biggest challenges of our time. At Demo Day 2026, discover how AI is accelerating scientific discovery and giving domestic workers more power in the workplace, how researchers are beginning to decode whale communication, and how innovative approaches to clean energy could transform how we generate power. Plus, explore how bookstores, art, and storytelling are helping us connect and find belonging. Watch the...

What Fei-Fei Li said

Written from the verified transcript and checked against it. Every figure links to the moment it was said.

Fei-Fei Li, co-founder and CEO of World Labs, discussed her journey from a childhood in China to Parsippany High School in New Jersey, where her math teacher Mr. Bob Sabella created a single-student calculus class during his lunch hour and mentored her as an immigrant. She credited Albert Einstein and her teacher as major influences. Li explained her focus on spatial intelligence, noting that visual perception predates language in evolution by over 500 million years and is fundamental to human cognition. She announced World Labs' latest model, Atlas, a pre-trained world model that can generate 3D worlds from text or images and reconstruct environments from photos, with applications in VFX and robotics. Li confirmed World Labs will join AMD, where she will serve as chief scientist while continuing to lead World Labs. She described becoming an entrepreneur as her most rebellious act, despite advice against it.

Key takeaways

  1. World Labs released Atlas, a first-of-its-kind world model that unifies 3D world generation and reconstruction from photos.
  2. World Labs will join AMD, with Li becoming AMD's chief scientist while continuing as World Labs CEO.
  3. Li cited visual perception as evolutionarily foundational, predating language by over 500 million years.
  4. Li's high school math teacher Mr. Sabella created a single-student calculus class for her during his lunch hour.
  5. Li called becoming an entrepreneur at World Labs her most rebellious act, despite friends warning it could damage her reputation.

Numbers and commitments

FigureWhat it refers toTypeAt
530 million years ago When visual perception emerged in the animal kingdom timeline 54:17

Chapters

  1. 0:00Early life and influences
  2. 48:43Mr. Sabella's mentorship
  3. 51:33Beginner's mind and curiosity
  4. 54:17Spatial intelligence rationale
  5. 57:54World Labs and Atlas model
  6. 59:31AMD acquisition and role
  7. 1:01:19Rapid fire questions

Questions asked in this interview

8
  1. 46:10And did you study Einstein in high school, or was this just your reading for pleasure?
  2. 48:43Did he give you extra work to do? How did he influence you?
  3. 53:33What drew you to spatial intelligence versus a large language model, the typical one?
  4. 57:14And how have you been surprised over the last year with the state of the technology and the progress?
  5. 1:01:19If you could witness any scientific discovery in history, which would you choose?
  6. 1:01:41No. So, which one is it?
  7. 1:02:02What's the most rebellious thing you've ever done?
  8. 1:03:55What's something in nature that still seems like magic to you?
Narrator 0:09 ↗
You're amazing. You know that, right? Look at you. Just look at you. Consider it. Contemplate it. Flip it and twist it. Reassociate it. Spidey sense it. Look at you. I mean, just look at you. Nurture it. Build it. Stand on it. Just look at you. Challenge us. Flabbergast us. Unleash us. Unlock us. Lift us. Lift us. Lift us. Look at you dare leap. Look at the future inside of you exploding out of you. You perfect, unstoppable human.
Loren 1:34 ↗
Hello. Hello, San Francisco. Welcome to our 10th annual Demo Day, if you can believe it. Amazing how lucky we are to be alive right now. We are living through extraordinary days. We're watching in real time as new technologies accelerate scientific discovery, transform industries, and open up and expand possibilities for human ingenuity and creativity. It's just remarkable. And we see the very same technologies can be used to divide and to harm and even to destroy. So this is what I want us all to remember and to keep inside of us. We get to choose how we want to be as humans. We get to choose how we care for and treat one another. We get to choose whether we build a more just and kind world. So, let's choose that.
You know, throughout history, humans have invented remarkable tools to deepen our connectivity and deepen our knowledge. Our innate curiosity has fueled our imagination and led to amazing scientific breakthroughs, new forms of artistic direction, and technologies that have already profoundly changed our world. Today you will meet people who are, thank goodness, questioning the status quo, who are experimenting with new ideas and approaching very difficult problems with originality and purpose. Their work expands what we believe and understand is possible. And you know, perhaps that is one of the great wonders of being human. With all that we've learned and all that we've accomplished, we're still capable of surprising ourselves. So, welcome to Demo Day. I am delighted now to bring to the stage my friend and colleague David Seamus, who will be our MC.
David Seamus 4:15 ↗
Oh my goodness, what a day. It's perfect. Loren, thank you. I'm David Seamus, a managing director at Emerson Collective, and I could not be more honored to be with you here today. We are so incredibly excited. This is our 10th Demo Day.
For a decade, we have brought together creators and innovators, people who have been building solutions to some of the biggest challenges of our time. Today will be no different. You're going to hear from entrepreneurs and scientists, artists and community leaders, people who are shaping our world for the better. And then we're going to close things out with a finale that I promise you, you will not soon forget.
And I'm going to make you one more promise. You are going to enjoy yourselves. So do not hold back. Cheer enthusiastically for every speaker who comes onto this stage to share their vision, their work, and more importantly, a little bit about who they are. So, let's get into it. In this first group of demos, you are going to hear from a visionary leader who's bringing new energy to Hollywood, a trailblazing organizer harnessing AI to empower America's domestic workers, a nuclear energy entrepreneur working to usher in a new era of clean and reliable power, and the marine biologist on an extraordinary and joyful quest. What a lineup. So, up first, please welcome Darren Walker.
Darren Walker 6:25 ↗
Good afternoon. Demo Day. 10 years of Demo Day. Congratulations to Loren and everyone at Emerson.
So a year ago at the ninth Demo Day, I was sitting in this audience just like you. I was working at a foundation in New York and thought I had things figured out. I knew what I would be doing next. And then I had an intervention and everything in my life changed. I thought I would be in New York working in philanthropy or working in investments, and this intervention led me to another place, emotionally, geographically. So I moved from New York to Los Angeles and I joined Anonymous Content, a company that is at the very heart of storytelling in this country. And I did this because I believe in storytelling. I believe that our country needs stories and storytellers more than we ever have because storytellers build the fabric of our culture. They create the soundtrack to tell us who we are. They remind us of our humanity. They expand our minds.
And so I found myself in Hollywood, a place I would never have thought I'd be. Living in a house and driving a car, which for this New Yorker is something I could never have imagined. And yet, when given the opportunity to lean in and think about a different future for myself, I jumped at the opportunity. So storytellers do great things. They help us as a people. And in Hollywood, the storytellers are going to save our industry. One of the major issues we have in Hollywood today is that people don't often see themselves and see how their futures can be changed. I remember the impact of a film with a deep story in it had on my life. I was a little kid in rural Texas and I saw Sidney Poitier in Guess Who's Coming to Dinner. Now, growing up poor and black in the South, I'd never seen a black man unapologetically fabulous, brilliant, erudite, cosmopolitan. I could imagine myself being taken from that little shack on that dirt road in Ames, Texas to another world. That story changed my life.
And when I joined Anonymous Content, one of the things we were grappling with and that the industry is grappling with is how to get young people back to theaters. In fact, many in Hollywood had given up on the idea that young people would want to see films and that they'd only want to see films on their phones. And then along came three films: Iron Long, Back Rooms, and Obsessions. These films changed the landscape. They grossed almost three quarters of a billion dollars and yet they cost $14 million to make. Everyone in Hollywood got excited about the idea of scaling this and figuring out how to, at the largest scale, efficiently finance these movies. And so I believe that they were focusing on the wrong thing. Not how, but why? Why are young people returning to theaters? Have you noticed that young people stand in long lines for seemingly trivial reasons? They stand in line all the time. My generation, if we see a line, we go in the other direction. Young people go for the line. And they go for the line because young people in America today crave community.
They are desperate for human-to-human contact. Generation Z came of age during COVID. Isolation marked their lives deeply, and so they crave human-to-human contact. And we know how powerful that is. I remember for myself going to a cinema in Austin, Texas in 1978 to see a film that was transformational for me too. It was called The Rocky Horror Picture Show. Now, for those of you who remember The Rocky Horror Picture Show, can you imagine doing the time warp in a room by yourself, looking at your computer screen and throwing popcorn? No. Because Richard O'Brien, who wrote that screenplay, he understood that people needed to be in that room and together experiencing joy and exhilaration and the excitement that comes with being seen and being heard and seeing yourself understood. And so we need in this country our storytellers to do the brave and courageous work that they do to hold the mirror up, to provoke us to open our minds and our hearts. So my call to action is that you all get in your car, get on the train, get on your bicycle, join some friends, and see cinema in America. I'll see you at the movies. Thank you.
Ai-jen Poo 13:32 ↗
When I first started organizing domestic workers almost 30 years ago, nannies traded information on park benches like, 'What is a fair wage?' or 'How do I ask for a sick day?' And the truth is today is not much different because without clear standards, abuse is actually still quite common and trusted information is precious. Domestic work is still some of the most undervalued and isolated work in our entire workforce today. There's no co-workers. There's no HR. There's no one to call if you've been underpaid or mistreated or just have a simple question about your job. That isolation pushed us to try to find new ways of connecting workers across millions of workplaces. We built the National Domestic Workers Alliance to help ensure that the millions of nannies, home care workers, and cleaners who care for our homes and families can care for their own, too. And we've used technology to help make that possible, from building a platform to give workers access to benefits to delivering emergency cash assistance during the pandemic.
And now with AI, we have the most powerful technology yet, technology that could potentially change all of our work in profound ways. And so that led us to ask the question: what if AI could be built by workers for workers? And what if AI could help make precarious, isolated jobs better? And what better workforce to start with than the workers whose work, the work of care, is more essential than ever, especially as the baby boom generation ages and we're all living longer. That's why we built Ask Ya. Ask Ya is an AI assistant for domestic workers that we built to help workers understand their rights and navigate difficult conversations and negotiations with their employers. We built it with a thousand domestic workers and we launched this year, and now thousands of workers are using it. And what makes it work is a combination of three things. The first is security. Workers built Ya to be a safe and affirming place to bring questions that you might be afraid to ask anywhere else. Questions like, 'How do I say no to my employer and not lose my job?' Workers even built it to protect their data security. They decided that all conversations will be automatically deleted after one week.
The second is dignity. Right now, workers have to navigate a confusing patchwork of local, state, and federal policies. Something as simple as overtime pay, depending on the state you live in, the kind of care you do, or whether you sleep under your employer's roof, might create a different set of protections for you. And Ya makes all of that really simple to understand in English and in Spanish whenever you need it. And the third is confidence. Just because you know that you deserve a raise does not necessarily mean that you have the confidence to ask for one. And what Ya does is it helps you rehearse conversations you need to have, produce sample messages, sample scripts, and there's even a person, a real-life person on our team ready to take your call if you need it. And what all of this means is that in our pilot, 93% of workers took action based on Ya's advice. And within six months, one in every four workers negotiated an increase in pay.
When workers have a hand in building technology that improves their lives, it changes their relationship to the AI debate. Instead of feeling like people who need protection from what's to come, they feel like people who have agency and power to shape the future. Ask Ya is an example of AI for good. But just imagine how it would benefit all of us if all of the people whose lives will be impacted by AI had a hand in building it. That is what's possible. Thank you.
Clay Sell 19:11 ↗
On December 2nd, 1942, Enrico Fermi and a team of scientists at the University of Chicago made history, achieving the first self-sustaining nuclear chain reaction and demonstrating that nuclear fission could be controlled and used. Just 11 years later, the US Navy took this same science and powered a submarine, and commercial nuclear power in the United States soon followed, generating 20% of our electricity and 60% of our clean power in the United States just a few decades later. This was the beginning of the first nuclear age. It showed the promise of nuclear power. It showed what it could do. But there was a problem. And the problem is in rare scenarios, the reactor could lose control. It could overheat. It could melt the fuel and release radioactive material, potentially threatening humans and the environment. So, we added systems. We added complexity. We added regulation. And the first nuclear age ground to a halt.
I saw the promise of nuclear power firsthand years later when I served as Deputy Secretary of Energy in the George W. Bush administration. I also saw how difficult it had become to design, to license, and deploy these reactors. A few attempts were made during that era, but they proved to be too expensive and the schedules were too long, and most, but not all, were cancelled. I still believed that the world needed nuclear power. Despite its challenges, it was still the safest form of power that we had and the best source of carbon-free, always-on power for the US economy that could be deployed at the greatest scale. So we saw an opportunity. What if we could invent a new way to do nuclear power? Not abandon it, but invent a new way. What if we could create a nuclear power plant that was physically impossible to melt down? Spoiler alert, we can. And that's exactly what we're doing at the company I lead called X Energy. So, what is our approach? We actually start with a gumball machine.
Not really, but it's a useful illustration. What we start with is our special sauce, our fuel, a large gumball that we call a TRISO-X pebble. Now, inside this pebble are 19,000 energy-dense uranium kernels wrapped in material that can never melt. Inside this pebble, the nuclear waste will remain permanently safe from the biosphere in which we live. The other interesting thing about our design is how we manage excess heat. Now, in traditional nuclear reactors, you have water systems and multiple redundant mechanical systems to move it around. But in our reactor, we just use the power of physics. We have designed the reactor in such a way that if it overheats, the nuclear chain reaction automatically and naturally shuts down. And if we were to lose helium out of the reactor, the heat would dissipate naturally, damaging none of the systems. It's an extraordinary, extraordinary technological development. So now we are on the cusp of a new nuclear generation that will unlock abundant energy and economic growth for decades to come. And in fact, just this week, on this Monday, the US Nuclear Regulatory Commission approved the design that I've described for you today for us to construct for an industrial customer in the state of Texas. Historic.
We're excited to get back to building. We're excited to create a future that some thought it was impossible to create. But the larger point I want to leave with you is this. When innovators like you encounter risks, your first instinct is often to add something. Add a system, add a process, add another layer of protection. And each of these can be justified in their own right, but over time they can create complexity that becomes its own obstacle. At X Energy, we believe the greatest innovation is not in what you add, but is in what you can design away. So what problem do you think you can solve in the very same way? Thank you.
David Gruber 25:28 ↗
Humans have been imagining contact with alien intelligences for a long time. In sci-fi, like ET and Arrival, we dream of encountering extraterrestrial life. I'm a marine biologist and the president and founder of an interspecies communication initiative called Project CETI, the Cetacean Translation Initiative, and we're named in homage of SETI, the Search for Extraterrestrial Intelligence, but instead of pointing our lenses outward, out into space, we're pointing them inward, into the ocean, towards sperm whales, using the latest in artificial intelligence, gentle robotics, linguistics, complex network science, natural language processing, and more.
So, why sperm whales? Why sperm whales? Sperm whales have these incredibly complex social lives. They live in multigenerational and matrilineal societies. So, they're female-led and consisting of grandmothers, mothers, and kids. And we've recently shown that they collaborate and they work extensively among kin and non-kin to elevate their babies at birth so they don't sink. They have the largest brains in the universe, a 20-pound brain. And they communicate with these like Morse code sequence of clicks known as codas. And we're just now beginning to understand the features of their communication system that were hidden in their voices. So here's some sounds.
The crazy part is that humans and whales actually share a common land-dwelling ancestor at 90 million years ago, Pakicetus, and perhaps now there's finally a moment where we could reconnect. So some of CETI's first findings were that they have a phonetic alphabet and their alphabet includes musical elements like tempo, rhythm, rubato, and from there, with the assistance of AI, we went from understanding that there were 21 distinct codas to 156, and using these musical elements they could combine them together almost to an infinite number of combinations, just like humans can. And working with linguists, we found features inside each click that resemble vowels and diphthongs. Essentially bringing the complexity level yet up to another bar. So what's been getting us really excited is building upon these findings. We've been developing our first translative whale acoustic models, WHAM. We're having fun naming them, too. Essentially, instead of predicting the next word, we're predicting the next click. And with the thousands of hours of this context of conversations that we've been recording, we're starting to learn more and more and more.
So the question I often get asked is like, Dave, how close are we? Which I think kind of misses the point of the vastness and beauty of what deeply listening to whales could offer humanity. But more importantly, this gives us the opportunity to imagine what we want interspecies communication to look like. And if you think of like the telescope allowed the unaided human eye to see out into the galaxies and the microscope to allow us to see into cells, AI and bioacoustics can allow us to hear and understand like we never could before. And I will be as bold to say is that the single reason AI was invented was to allow us to connect deeper to the millions of species that we share the natural world with. In the 1997 movie Contact, Jodie Foster, she encounters this breathtaking celestial event and she says they should have sent a poet. It's so beautiful, highlighting the disconnect between science and this deep awe of human experience of being alive. At CETI, we aim to bridge that disconnect. What could a deeper interspecies understanding bring us? So, I leave you with that. What question would you ask? What would you want to understand from whales when this moment to really deeply understand them comes? Thank you.
David Seamus 30:04 ↗
David, I told everyone that was going to be magical. So, you have a room filled with about 700 baby whales who want to dig deeper into your work and refine their whalish. And so, why don't you tell them where to find your work and how to support you?
David Gruber 30:24 ↗
Oh, thank you, David. Well, we just started our Listen to the Whale campaign in collaboration with National Geographic. So, it's listen.projectceti.org, and we have our deep archive where we're trying to get 8.3 billion people to go in there and to say what they would want to learn from whales. So, really kind of like take this moment and say, asking everybody before we reach it, and what they would want to learn from them. So, thank you.
David Seamus 30:51 ↗
David, thank you. 700 baby whales down, 8.3 billion to go.
David Gruber, thank you again. And Darren and Ai-jen and Clay, what incredible work from our first round of speakers. Let's give another round of applause to these folks.
And I would be remiss if I didn't thank Ai-jen for actually returning to this stage. At our very first Demo Day in 2017, she talked about the importance of telling the stories of listening to domestic workers as a form of advocacy. So, it's deeply moving to me and to all of us that these workers now can use this technology as a tool to advocate for themselves in a way that was never possible before. And in that same vein of deep listening, we're in for another really special treat. I am delighted to introduce an actor and playwright who's created an innovative form of documentary-style theater based on the spoken words of the people she meets. Please welcome Anna Deavere Smith.
Anna Deavere Smith 32:28 ↗
We are in the presence of Jack Whitten's majestic painting, Autotopolis. Autotopolis, he wrote, is a made-up word that means no place from Greek roots. A nottopos, place, polus, city. He jotted on a yellow pad. It's an imaginary city constructed out of elements from anywhere. A borderless city built from the uprooted, ungrounded, and nomadic destinies of old. A fluid identity. Whitten proposes that we must create our own belongingness. The fragments in the painting find their way to, if not connectedness, a relationship, a relationship with that which might not fit. The painting has become the north star of my Emerson Collective project, the Deavere Smith Performance Labs, aka Pipeline Girls. Since 2021, I've conducted a performance lab with a group of young women, watching them progress from girlhood to womanhood. They arrive each year out of the streets of Baltimore, the affluence of Orange County, California, the salmon fishing riverside of the Yurok reservation in Northern California, the bison ranges of the Blackfeet Nation in Montana. They find their way towards each other not by trying to meld into sameness, but by aggressively reaching across the divide.
This past summer, I asked each participant to identify the person in the group who they would say was the most different from them. What I call the not you, the not me. Inspired by a conversation with Emerson Collective's Alex Simon, I then worked with each girl to perform their not me. In the process, I found a not me for myself. To enrich my inquiry, I had opened the lens to include family members. Brittany Davis is the sister of one of our participants, Sariah. Sariah spent a part of her childhood in a Baltimore homeless shelter. She is now a junior at Johns Hopkins University with a free ride. She considers Brittany, her older sister, her rock. And so for Demo Day, I put together some verbatim excerpts from an interview that I conducted of Brittany. Brittany's story exemplifies that hard work that is required to create one's own belongingness, and she's one of my not me. I hope you won't mind, but Cody, who's five, and Charlie, who's seven, might come out on the stage in the middle of my performance. So this is Brittany Davis, word for word from her interview.
Brittany Davis 35:25 ↗
Sariah is in Japan. Amazing. She's living her best life out there. My home has always been a second home to Sariah. She's my sister. She's my mother's daughter. We don't share the same father. So when my mother had... So to be funny, well, people didn't find it funny at the time, but that's my story. So when my mother had me, my mother, and my oldest sister was all pregnant at the same time. I wasn't supposed to be pregnant. I got pregnant at 13. In a second. I'm on an interview. Give me a second. I'll be in a second. Thank you. I'm sorry, y'all. They out of school. I'm trying to multitask. So, shut the door.
The most challenging thing was I was pregnant at 13. I wasn't an adult, right? I didn't have any income. How I'm going to feed my baby? And so I said to my family, 'Well, I'm a breastfeed.' And it was just like, 'Oh, you don't want to do that. You're young. You don't know what that entails.' So I did a lot of research. I seen people do it, right? And I did it, right? I did it. And it was just one less thing somebody didn't have to worry about providing for my child. I tried my best to keep the burden off my family. If it was going to somebody's house and say, 'Hey, let me help you clean your house.' And then I used to braid. I've been braiding ever since I was young. I braid hair. Or I go to a neighbor house that I know. 'Hey, you want me to take your trash out? You want me to pull the weeds out your yard?' And I wouldn't say that anybody necessarily protected me. I protected myself, I guess.
And when my mother found out I was pregnant, her and my aunt was like, my aunt on my father's side was like, 'Well, we're going to take Shawan,' they call me Shaw, 'we're going to take her to New York.' And they took me to New York. They took me to Planned Parenthood, and they take you in a room, they ask you a bunch of questions. How you feel about it? Lady asked me, 'Did I want to do it?' Now, from my understanding, if you say no, no means no to them. And then if you say no, they give you an option to have an adoption. And I didn't know what I was getting into, but I'm not giving y'all my baby. This is my baby. So, it's a no. My family could have kicked me out. I don't... it didn't... I didn't care. This is my... this is my... and no at 13. No.
Having a baby, being a baby, I took a lot. I took a lot of hits. I took a lot of hurtful things. I took a lot. But I took it, right? I took it. Growing up at the church, it was hard. I mean, just maneuvering through the world. People would see me on the bus, they would talk about me. I remember people, family members, I remember family, remember one of my aunts was like, 'All she going to ever be good for is laying on her back.' Right? Where did that come from? 'Cause I wasn't a fast girl in a sense, right? I mean, yeah, I made a mistake, but I made a mistake because I didn't know better. Nobody told me that once you get your period or what sex even was, right? I didn't know what sex was when I was 13. I knew I did something. But nobody talked about it back then. That wasn't a thing. You was a child and you stayed in a child place. Y'all two stop. Please. In a second when I finish, play nice.
People say some harsh things to me. Call me a... a... but I'm not a grudge holder. I don't hold people accountable for the things that they say. They don't know no better. It doesn't resonate with my soul. And people always ask me, 'You never felt depressed in your whole life?' And I'm like, 'No, I never felt depressed in my whole life. And what's this thing everybody talking about now? Overstimulated.' I have been pushing through challenges my whole life. The word can't is not allowed in my house. And yes, I have a story. I was 13 with a baby. And it's hard. You might cry, tear or two, but it's a moment. Have your moment. Get up, brush it off, and keep it moving. Y'all better not be fighting.
David Seamus 40:39 ↗
Amazing. Amazing. Once again, thank you, Anna Deavere Smith.
No matter how many times you see someone at the height of their craft in their work, it's still astonishing to be able to watch something like you just saw where another human being can capture the emotion and the stories of someone else and convey it in a way that just moves you. That's the beauty of Demo Day for all of you to continue the journey both with Anna and with Darren, Ai-jen, and Hans Ulrich. And now it's time for our first conversation, and it's with one of the most influential leaders in artificial intelligence. So please welcome Loren back to the stage, joined by Dr. Fei-Fei Li, the co-founder and CEO of World Labs.
Loren 41:53 ↗
Hi, Fei.
Fei-Fei Li 41:54 ↗
Hi, Loren.
Loren 41:55 ↗
Welcome to Demo Day.
Fei-Fei Li 41:56 ↗
I know this is exciting. Hi, guys.
Loren 42:01 ↗
So as David said, Fei-Fei Li has helped shape the modern era of artificial intelligence, and she is the co-founding director of Stanford's Institute for Human-Centered AI, which is a really important distinction and obviously a great influence to what came next. While she was there, she established a very powerful premise, which is this: technical progress and human values must move forward together. This is at the core of what she believes and how she works, and in a way, I think other parts of the industry have come along right behind you, but you were there 10 years ago. And today, through her company World Labs, she's developing AI systems that are capable of understanding and generating three-dimensional worlds. So her work is called spatial intelligence, and it opens up new possibilities in science, creativity, and industry. Fei champions a vision of AI defined not only by its new and astounding capabilities, but also by its potential to expand what human beings can imagine and accomplish. So Fei, thank you so very much for coming here today at Demo Day.
Fei-Fei Li 43:30 ↗
Thank you for inviting me. Thank you.
Loren 43:36 ↗
And I wonder, Fei, if you can talk to us a little bit about how you got to where you are. So let's start at your upbringing in China, and then you moved to the great state of New Jersey for high school. And a lot of your early scientific awareness and openness was shaped by your father, who had deep curiosity about nature. And you've written about him in your book, and you've spoken about him. And I wonder if you can share with us some of the other major influences on your education, on your scientific curiosity.
Fei-Fei Li 44:19 ↗
Yeah. Well, thank you, Loren. I think like many of you, my own journey started with curiosity. You know, just this girl in China who fell in love with physics when I was probably a preteen, and my family and I moved to New Jersey. We went to a public high school called Parsippany High School. And despite this drastic change of immigration life, despite the fact that I had to learn a new language and settle in as an immigrant, and you know, the early years of immigrant life is not necessarily all rosy, my love for physics never settled. In fact, it just grew. And so you asked me who are my influencers in that early transformative years. Well, there are two people. One person is very dear to me in real life. Another person has long passed away. You would never know how much he influenced my life. That one was Albert Einstein. And especially being a New Jersey girl, having Princeton campus is where I spent the rest of his immigration life. It was so special to me. I just... he obviously, we all know him as one of the greatest scientists in human history, but he was also a humanist. His curiosity of nature, science, as well as his compassion for humanity, had deeply impacted me throughout my career.
Loren 46:10 ↗
How interesting. And did you study Einstein in high school, or was this just your reading for pleasure?
Fei-Fei Li 46:18 ↗
Well, I'm not going to lie, I was dreaming about relativity when I was in high school.
Loren 46:24 ↗
Same. Yes. Okay. No, it's really... I was obsessed with physics.
Fei-Fei Li 46:32 ↗
Yeah, it was a lot. I studied a lot both in class as well as by myself, and I continue to read about Einstein's writing and just pick up bits and pieces. And you know, he was a musician. He has all these letters where he deeply, deeply cared about what's going on in the world. He's a Jew and all that. So anyway, that's one person. Another person who deeply influenced me, and I think just given this audience, it's very important to acknowledge people like this, was my high school math teacher, Mr. Bob Sabella. Yeah, it's, you know, I was a new American, and even after so many decades, I look back at my own journey, these high school, public school teachers are truly the unsung heroes of ours.
Loren 47:36 ↗
Yes. Yes.
Fei-Fei Li 47:42 ↗
It's especially just to tie it to today, right? The whole world talks about AI, and we see the big transformations. But I just often think about these most important people in our society are the teachers and the students. And here we are in Silicon Valley talking about the most macroscopic topics. And the kind of money we talk about comes in the unit of millions, billions, or even trillions. But really, the teachers in everyday classroom, teaching our inner-city kids, teaching our rural students, teaching immigrant kids, they are some of the most important people we need to support and remember in this.
Loren 48:32 ↗
We agree with you. Totally agree with you.
And what was his name? Mr. Sabella?
Fei-Fei Li 48:43 ↗
Mr. Sabella.
Loren 48:43 ↗
What did he do that helped encourage you? Did you meet with him after class? Did he give you extra work to do? How did he influence you?
Fei-Fei Li 48:56 ↗
Yes. So he was my math teacher, and he was a very good math teacher, very tough. I had moments of weakness that he would be very tough and had this kind of tough love. But quickly I got placed out of the calculus AB, and Parsippany was not a well-resourced community. We don't have multivariate calculus. I couldn't... we don't have community college for me to take advanced math. So he created a math class for me as a single student during his lunch hour.
Loren 49:38 ↗
Wow.
Fei-Fei Li 49:40 ↗
But here he and his family, Jing Sabella is also a public school teacher, they really took me in as this immigrant kid who is just... in the early years of her journey as an immigrant who was lost, who was frustrated, who needed guidance. And they really took me in as their third kid. But the one thing I want to say about Bob Sabella, and I know there are so many teachers like this in America, is he didn't just do this for me. There are many, many what we call ESL kids, English as second language kids, and he paid attention to all these kids. He became an advocate. He became a mentor. He became a friend, and he just went above and beyond for so many students. And the reason I know that is after I graduated, we became lifelong friends, and he would share some of his journals with me, and that's his personal journal every week, every month, full of stories of students in Parsippany High School.
Loren 50:53 ↗
Wow.
Fei-Fei Li 50:53 ↗
So these are, you know, it's just such a beautiful... he's very special, but he's also one of many.
Loren 51:00 ↗
Yes, it's true. That's a beautiful story. I love it. And I have personal pride that it happened in New Jersey, but it could have happened anywhere in America. You strike me as someone who has preserved almost a childlike sense of wonder. You have a fully active imagination and sense of wonder. And I wonder if you've ever given thought to what children understand about the world that adults often forget.
Fei-Fei Li 51:33 ↗
Wow. You and I both know the concept of beginner's mind. You know, I also read it from Steve's writing.
Loren 51:42 ↗
Yes.
Fei-Fei Li 51:43 ↗
And I think that is what children teach us. The way they approach the world is with a beginner's mind, is that they... you know, I also did neuroscience when I was in PhD time in addition to AI and just trying to understand human cognition. It is fascinating how our brain evolves over evolution as well as develops through a course of a lifetime. I think that, you know, we humans in the early years of our lives, our sensory, our perception, our cognitive inquisitiveness of the world is just so open. And in a way, we're also more free from maybe the daily burdens of life. And that state is so precious. It allows you to fully experience the world and also ask questions. Yes. And as a scientist, it's hard for me to describe why I preserve that, but I think as a scientist, that passion for curiosity is one of the most important things. And I feel very lucky every day I wake up. That is still the singular most important energy that drives me and drives my work. And I also get to work with students, you know, who in my eyes are getting younger and younger because I'm getting older and older. But having a career working with young students or engineers or researchers allows me to also share that curiosity and passion on a daily basis.
Loren 53:33 ↗
Yeah. When you were getting your PhD and you could have gone in a couple of different directions, you ended up obviously being fascinated by spatial intelligence. And so this is something that Fei has been working on for 30 years. For a lot of the world, AI is new and everyone's grappling with it. But you've been thinking about it and working on it and the overlap with neuroscience for a long, long time. What drew you to spatial intelligence versus a large language model, the typical one?
Fei-Fei Li 54:17 ↗
Yeah. So the word spatial intelligence, or sometimes we call visual, but now the combination between spatial and physical as well as robotics is all, you know, the progress of the technology is so fast, it's all converging. But there are two things that really make me love this whole world of visual, spatial, physical intelligence. First of all, if you look at evolution, we in the animal kingdom had perception and are sensing the world in a visual way more than 530 million years ago. Language to humans came a lot later. If we're very generous, proto-language started maybe half a million years ago. So perception of the world is a much, much more primordial, foundational capability for animal nervous systems. And not only that, fossil studies have suggested it's also the driving force of evolution itself. Because with the capability to perceive and to understand the physical world, to navigate, to interact with, animals developed much more active life in terms of being preyed on, being the predator, reproductive pressures and all that. So the causal connection or the link between perception, action, and an evolutionary progress of intelligence is extremely tight. So that fascinated me.
Also for us as individual humans, babies start preverbal. We start to understand the world through seeing and touching. And also even today a large fraction of our cortical activities and functions are devoted for visual. We not only communicate and do work through languages. So much of our daily life from the moment you wake up from your bed, get yourself down to the kitchen, prepare breakfast, drive to work, you know, everything you do in between, drive back, hang out with friends, watch sports, do whatever hobby you have. So much is visual and physical. So this continues to fascinate me is how do we understand this part of intelligence and how do we enable machines to do that so that it can help humans.
Loren 57:14 ↗
Yeah, I'm very glad we're listening to you after having listened to David Gruber and understand the richness and sophistication, and that we're only just touching the surface of with sperm whales, let alone any other species. So what you're saying is really resonant. So talk to us a little bit about where World Labs is now. What is the state of the technology? What are some of the challenges that are still in front of you? And how have you been surprised over the last year with the state of the technology and the progress?
Fei-Fei Li 57:54 ↗
Yeah. So just to give you a little bit of a history, World Labs was founded about two and a half years ago, almost three years ago now, with a group of technologists including my former students. And at that time we believed that the ability to allow AI to generate pixels and reproduce space and simulate the physical movements of objects in space is very important for VFX, for designers, as well as for robotics. As World Labs grew, the state of the technology also grew. So, as of yesterday, we actually made public our latest model called Atlas. And Atlas is this pre-trained world model that can generate worlds. You can literally prompt it with a sentence or with an image and it will generate a 3D world that can allow this kind of interactive experiences. But it also can allow you to reconstruct the world if you give Atlas a few photos of a realistic space or environment that you want to reproduce, especially for storytelling or for robotics. And Atlas is this first-of-its-kind world model that unifies the capability of generation as well as reconstruction. And we're showing applications in VFX as well as robotics.
Of course, the LLMs are now becoming very multimodal. I'm fully aware even Astra, Fable are demonstrating incredible capability of driving computer software such as Blender or Unreal Engine to create 3D worlds. So it's really interesting to see a different technological take trying to do similar things. I'm not going to get into the details. What Atlas can do today that combines generation and reconstruction of real world is still very unique, and we do hope that our users and developers get to play with that and use that. And also, you already announced that World Labs will be joining AMD very soon as part of the AMD family, as AMD as a company is transforming itself from a chip company to an AI platform company, and I'll be leading both World Labs and be AMD's chief scientist, and I'm really excited by this new chapter as well as after we close.
Loren 1:00:48 ↗
Yes.
Fei-Fei Li 1:00:48 ↗
To deepen...
Loren 1:00:49 ↗
That's very exciting.
Fei-Fei Li 1:00:50 ↗
To deepen the world modeling, spatial intelligence work, but also to contribute more to the open ecosystem, which would be really fun and exciting.
Loren 1:01:03 ↗
Oh, I'm sure a lot of people can't wait to play with Atlas.
Fei-Fei Li 1:01:07 ↗
I want to see it.
Loren 1:01:08 ↗
Okay, it's time for our rapid fire round of questions.
Fei-Fei Li 1:01:14 ↗
No overthinking it. Fei, Fei. All right. I'm scared.
Loren 1:01:19 ↗
Okay. If you could witness any scientific discovery in history, which would you choose?
Fei-Fei Li 1:01:27 ↗
Well, I already gave a...
Loren 1:01:30 ↗
Well, let's just... It has to be Einstein.
Fei-Fei Li 1:01:33 ↗
I'm debating between special relativity and general relativity.
Loren 1:01:41 ↗
Didn't one of them happen on a trolley when you had an insight? No. So, which one is it?
Fei-Fei Li 1:01:49 ↗
I'll say I'll go for general relativity because this concept of space, time, and mass is just so beautiful. It's a nerd thing.
Loren 1:02:02 ↗
Okay. What's the most rebellious thing you've ever done?
Fei-Fei Li 1:02:09 ↗
Oh, wow. That's a good one.

60 more exchanges in this transcript

Sign in free to read the rest of this interview. No card required.

Sign in to read the full transcript

Cite this transcript

APA, MLA, BibTeX
APA

Li, F. (2026, November 16). Emerson Collective Demo Day 2026 | Full Program [Interview transcript]. Emerson Collective. CEOInterviews.AI. https://ceointerviews.ai/interview/3303084/

MLA

Fei-Fei Li. "Emerson Collective Demo Day 2026 | Full Program." Emerson Collective, 16 Nov. 2026. Transcript, CEOInterviews.AI, https://ceointerviews.ai/interview/3303084/.

BibTeX
@misc{li2026_3303084,
  author       = {Fei-Fei Li},
  title        = {Emerson Collective Demo Day 2026 | Full Program},
  howpublished = {Interview transcript, Emerson Collective. CEOInterviews.AI},
  year         = {2026},
  month        = {nov},
  url          = {https://ceointerviews.ai/interview/3303084/},
  note         = {Speaker-attributed transcript with timestamps}
}