Fei-Fei Li

Fei-Fei Li
Born	Li Fei-Fei 7/3/1976
Birthplace	Beijing, China
Nationality	Chinese-American
Occupation	Computer scientist, artificial intelligence researcher
Employer	Stanford University
Known for	ImageNet, AI research, machine learning, computer vision
Education	Princeton University (B.A.) California Institute of Technology (Ph.D.)
Alma mater	Princeton University California Institute of Technology
Awards	IEEE Fellow; ACM Fellow; National Academy of Engineering; National Academy of Sciences; American Academy of Arts and Sciences
Website	profiles.stanford.edu/fei-fei-li

Fei-Fei Li (born July 3, 1976) is a Chinese-American computer scientist and artificial intelligence researcher whose work in computer vision and machine learning has reshaped how researchers and technologists approach the challenge of teaching machines to see and understand the visual world. Born in Beijing, China, and raised partly in Chengdu, Li immigrated to the United States as a teenager, eventually earning her doctorate from the California Institute of Technology and building a distinguished academic career at Stanford University, where she holds the Sequoia Capital Professorship in the Stanford School of Engineering. She is best known as the creator of ImageNet, a large-scale visual database that became a foundational resource for the deep learning revolution of the 2010s. Li served as Director of the Stanford Artificial Intelligence Laboratory and spent time as a vice president at Google, where she led Google Cloud artificial intelligence and machine learning efforts. Her work sits at the intersection of technical research and public advocacy, and she has spoken and written extensively about the societal dimensions of artificial intelligence, including questions of ethics, diversity, and access. She is the co-founder of AI4ALL, a nonprofit organization dedicated to broadening participation in artificial intelligence education.

---

Early Life

Fei-Fei Li was born on July 3, 1976, in Beijing, China, and spent her childhood in Chengdu, in the Sichuan province. She has described growing up in a family that valued intellectual curiosity, with her father working as an engineer. In 1992, when Li was sixteen years old, her family immigrated to the United States, settling in Parsippany, New Jersey.

The transition was not without difficulty. Li's family arrived with limited financial resources, and her parents worked demanding jobs — her mother in a factory and her father repairing watches and cameras — while Li attended high school and worked part-time to help support the household. She has recalled in interviews that she took on a job at a local dry cleaner during her teenage years to contribute to the family's income.

Despite the economic pressures of her family's circumstances, Li excelled academically. She developed a strong interest in mathematics and the sciences during high school, a path that would lead her toward physics and, ultimately, to the study of artificial intelligence.

---

Education

Li enrolled at Princeton University, where she pursued a degree in physics. She graduated with a Bachelor of Arts in physics in 1999, completing her undergraduate thesis on theoretical aspects of condensed matter physics. Her time at Princeton marked a turning point in her intellectual interests: exposure to work in neuroscience and the computational questions surrounding human perception began drawing her attention toward the problem of how the brain — and potentially machines — process visual information.

After Princeton, Li briefly considered other paths, including a period during which she operated a small laundry business in the summers to help fund her graduate school plans. She was admitted to the doctoral program at the California Institute of Technology (Caltech), where she studied under the supervision of Pietro Perona, a prominent researcher in visual recognition and computer vision. Li completed her Ph.D. in 2005, with her doctoral research focused on computational models of visual object recognition — work that laid intellectual groundwork for her later large-scale data efforts.

---

Career

Early Academic Work

Following her doctorate, Li joined the faculty of the University of Illinois Urbana-Champaign before moving to Princeton University as an assistant professor. She subsequently joined Stanford University, where she became a core member of the computer science faculty and has remained ever since. At Stanford, Li pursued research questions at the boundary of machine learning, computer vision, and cognitive neuroscience — asking not only how machines might recognize objects in images but also how the structure of human visual cognition could inform the design of artificial systems.

ImageNet

Li's most consequential contribution to the field is ImageNet, a large-scale image database she conceived and built beginning around 2006 and launched publicly in 2009. Motivated by a conviction that the field of computer vision was constrained by the small size and limited diversity of its training datasets, Li argued that machines needed exposure to the sheer volume and variety of images that humans encounter over a lifetime of visual experience. ImageNet, assembled with significant help from Amazon Mechanical Turk crowdsourcing and a large team of students and collaborators, ultimately contained more than fourteen million labeled images organized into more than twenty thousand categories based on the WordNet hierarchy.

Beginning in 2010, Li and her collaborators organized the ImageNet Large Scale Visual Recognition Challenge (ILSVRC), an annual competition inviting research teams worldwide to test their image recognition algorithms against the ImageNet dataset. The competition became a central proving ground for the field. In 2012, the entry from Alex Krizhevsky, Ilya Sutskever, and Geoffrey Hinton — the deep convolutional neural network known as AlexNet — produced a dramatic improvement in recognition accuracy compared to prior approaches, an event that is widely credited with accelerating the broader adoption of deep learning techniques across artificial intelligence research and industry.

The ImageNet challenge helped establish benchmark-driven evaluation as a norm in machine learning research and provided a dataset that researchers have continued to use as a reference point for measuring progress in computer vision for more than a decade after its creation.

Stanford AI Lab and HAI

Li served as Director of the Stanford Artificial Intelligence Laboratory (SAIL) from 2013 to 2018, overseeing one of the most active and prominent AI research groups in the world during a period of rapid change in the field. Under her directorship, SAIL engaged with a broad range of research areas, from robotics and natural language processing to healthcare applications of machine learning.

In 2019, Li co-founded the Stanford Institute for Human-Centered Artificial Intelligence (HAI) alongside John Etchemendy, the former Provost of Stanford. HAI was established with the stated mission of advancing research, education, and policy work on artificial intelligence in ways that account for human needs and societal consequences. Li has served as co-director of HAI since its founding.

Google Cloud

From 2017 to 2018, Li took a leave of absence from Stanford to serve as Vice President of Artificial Intelligence at Google and Chief Scientist of Google Cloud AI. In this role, she focused on developing Google Cloud's machine learning products and making AI tools more accessible to businesses and developers outside of major technology companies. Her tenure at Google also drew attention to broader questions about diversity in technology leadership; Li has spoken about the importance of increasing the representation of women and underrepresented minorities in AI research and development.

World Labs

In 2023, Li co-founded World Labs, an artificial intelligence startup focused on spatial intelligence — the capacity of AI systems to understand and reason about three-dimensional physical environments. The company raised significant venture capital funding and attracted attention as an effort to move AI research beyond language and two-dimensional image recognition toward richer models of the physical world.

AI4ALL

Li co-founded AI4ALL in 2017 alongside Olga Russakovsky and Rick Sommer, with the goal of increasing diversity and inclusion in artificial intelligence education. Originally launched as a summer program at Stanford for high school students from underrepresented backgrounds, AI4ALL expanded to include programs at multiple universities across the United States. The organization operates on the premise that broadening the range of people who develop and shape AI will improve the technology's outcomes for society.

Advocacy and Policy

Li has testified before the United States Congress on the subject of artificial intelligence policy, including testimony to the United States Senate addressing questions of AI safety, regulation, and the United States' role in global AI competition. She has consistently argued that effective AI policy should be informed by scientific expertise and that the governance of AI should involve a diverse array of voices from beyond the technology industry itself.

Li published a memoir, The Worlds I See: Curiosity, Exploration, and Discovery at the Dawn of AI, in November 2023. The book recounts her personal history alongside the intellectual and institutional history of modern AI research.

---

Personal Life

Fei-Fei Li lives in the San Francisco Bay Area. She is married to Silvio Savarese, also a computer vision researcher and a professor at Stanford University. The couple have children together. Li has spoken in public forums about the experience of navigating an academic and technology career as an immigrant and as a woman in a male-dominated field, and she has described the influence of her family's immigration story on her approach to her work and her understanding of what is at stake in the development of artificial intelligence.

---

Recognition

Li has received numerous professional distinctions reflecting her contributions to computer science and artificial intelligence. She was elected to the National Academy of Engineering and the National Academy of Sciences, two of the most selective honorary bodies in American science and engineering. She is also a member of the American Academy of Arts and Sciences.

Li is a Fellow of the Institute of Electrical and Electronics Engineers (IEEE) and a Fellow of the Association for Computing Machinery (ACM). She has received honorary degrees from institutions including Princeton University.

In 2015, Li delivered a TED Talk titled "How We're Teaching Computers to Understand Pictures," which has accumulated tens of millions of views and introduced the significance of her ImageNet work to a broad public audience.

She has appeared on lists of influential figures in technology and science published by outlets including Time magazine and Forbes.

---

Legacy

ImageNet's influence on the trajectory of contemporary artificial intelligence is extensively documented in the academic and journalistic literature covering the development of deep learning. The dataset and the challenge associated with it are credited by researchers including Yann LeCun, Geoffrey Hinton, and Yoshua Bengio — the three recipients of the 2018 Turing Award for their foundational contributions to deep learning — as having played a crucial role in validating and accelerating the practical application of neural network methods.

Li's career reflects a sustained effort to connect technical research in machine intelligence with broader questions about how societies develop, regulate, and share access to powerful technologies. Her institutional contributions — building SAIL, co-founding HAI, and establishing AI4ALL — represent an attempt to ensure that the structures of academic AI research remain attentive to questions of inclusion, ethics, and public benefit. Through both her scientific output and her public role, Li has occupied a position at the center of one of the most consequential technological transformations of the early twenty-first century.

---

References

---

Categories

---