Yann André LeCun (originally spelled Le Cun; English pronunciation: /ləˈkʌn/ lə-KUN, French: [ləkœ̃] or [ləkɛ̃]) is a French-American computer scientist specializing in artificial intelligence, machine learning, computer vision, mobile robotics, and computational neuroscience.^{[2,1,45,46,44,50,55,6,8,5,65,103,104,102,165,174,173,3,12]} Sources differ regarding his exact date of birth, claiming July 8, 1960, January 1, 1960, or June 6, 1960.^{[2,1,50,55,6,8,5,65,174,173,3,12]} Accounts also vary on his place of birth, reporting it as Soisy-sous-Montmorency in Val-d'Oise, Paris, or Soissons, France.^{[55,103,104,102,1,2,12]} His family surname originates from the Guingamp region of northern Brittany from the old Breton form "Le Cunff" (meaning "a good man"), which he later changed from two words ("Le Cun") to a single word ("LeCun") after moving to the United States because "Le" was frequently mistaken for a middle name; during a 2017 lecture in China, he also adopted the Chinese name Yang Likun (楊立昆 / 杨立昆), while in Albanian he is written as Jan Andre LeKën.^{[166,102,167,12,13]} He developed an early passion for mathematics and science, going on to earn a bachelor's degree in Applied Mathematics from Université Pierre et Marie Curie (UPMC) in Paris and a degree from the École Supérieure d'Ingénieurs en Électronique et Électrotechnique (ESIEE), before completing his PhD in Computer Science at UPMC in 1987.^[43] Following his doctorate, LeCun worked as a postdoctoral researcher at the University of Toronto's Department of Computer Science, joined AT&T Bell Laboratories in 1988 where he developed early machine learning and convolutional neural network methods, and moved in 1996 to the AT&T Labs research center to work on DjVu image compression technology.^[43]

In 2003, LeCun became a professor at New York University (NYU), where he served as founding director of the NYU Center for Data Science and holds the Jacob T. Schwartz / Silver Professorship of Computer Science at the Courant Institute of Mathematical Sciences, affiliated with the Center for Data Science, the Center for Neural Science, and the Electrical and Computer Engineering Department.^{[3,2,97,98,54,6,19,43,42,172,171,176,175,1,50]} In December 2013, he joined Facebook (now Meta) as Director of Facebook AI Research (FAIR) in New York, subsequently serving as Vice President and Chief AI Scientist at Meta until late 2025.^{[3,2,43,11,97,98,54,103,104,102,6,19,42,172,171,177,195,1,225,14,44,226]} In late 2025, LeCun left Meta to co-found Advanced Machine Intelligence Labs (AMI Labs), an AI startup focused on understanding and predicting the real world.^{[53,11,106,225,14,44,226]} LeCun has expressed skepticism toward current Large Language Models (LLMs), arguing that they lack genuine understanding of the physical world, and instead advocates for Joint Embedding Predictive Architecture (JEPA) models capable of constructing internal world representations.

LeCun is widely recognized as a founding father of convolutional neural networks (CNNs), notably developing the LeNet architecture for optical character recognition and computer vision.^{[5,6,4,42,3,14,66,12,94,165,179,180,178,221,8,19,74,11,17]} His CNN framework introduced specific architectural principles designed to handle object translation invariance in visual data, which later proved equally effective for processing linguistic sequences.^[66] Alongside Léon Bottou and Patrick Haffner, he co-created the DjVu image compression technology.^[71,6,5,165] In addition, LeCun and Léon Bottou co-developed the Lush object-oriented programming language.^{[5,181,165,6]}

In 2018, LeCun was awarded the ACM A.M. Turing Award (often referred to as the "Nobel Prize of Computing") alongside Yoshua Bengio and Geoffrey Hinton for their pioneering contributions to deep learning.^{[7,35,6,12,54,67,94,9,10,8,11,13,36,3,42,50,39,2,224]} Together with Bengio and Hinton (and occasionally Jürgen Schmidhuber), LeCun is widely celebrated as one of the "Godfathers of AI" and "Godfathers of Deep Learning."^{[8,7,9,10,11,13,14,3,66,12,182,6,67,39,2,224]} His additional honors include the IEEE Emanuel Piore Award, the 2015 IEEE/RSE Wolfson James Clerk Maxwell Award, the Princess of Asturias Award, the VinFuture Prize, and the Queen Elizabeth Prize for Engineering.^{[67,39,2,12,224]}

Yann LeCun's surname was originally spelled Le Cun as two separate words, originating from his ancestors in the Guingamp region of northern Brittany, France, where the name derives from the Old Breton Le Cunff, meaning "good man".^{[102,13,166,220]} After moving to the United States, many people mistakenly thought "Le" was a middle name, leading him to change the spelling of his surname to "LeCun" in the 1980s and 1990s.^{[102,13,166,220]} Regarding his Chinese name, his original translated name was Yang Leqiun.^[220] However, during a lecture in China in 2017, he provided and used his official Chinese name, Yang Likun.^{[102,13,166,220]}

Yann André Le Cun was born on July 8, 1960, in Soisy-sous-Montmorency, a suburb of Paris, France.^{[47,13,14,11]} His surname, originally spelled Le Cun, derives from the Old Breton form Le Cunff—meaning "elegant person"—which originates from the Guingamp region in northern Brittany.^[47] His first name "Yann" is the Breton equivalent of "Jean" or "John".^{[14,3,183,223]} He holds American citizenship and has three sons, a brother who works at Google outside the field of AI, and an older sister who is a professor, also not involved in AI.^{[17,122,44,45,69]}

In 1983, LeCun earned a Diplôme d'Ingénieur (engineering degree) from ESIEE Paris (École supérieure d’ingénieurs en électrotechnique et électronique).^{[14,56,68,13,184]} In 1987, he completed his PhD in computer science at Université Pierre et Marie Curie (now Sorbonne University) with a doctoral thesis titled "Modèles Connexionnistes de l'apprentissage".^{[14,56,68,94,13,184]} During his doctoral studies, he proposed an early prototype form of the backpropagation learning algorithm for artificial neural networks.^{[14,68,13,3,184]} From 1987 to 1988, he completed a year of postdoctoral research under the supervision of Geoffrey Hinton at the University of Toronto.^[44,12,14]

In 1988, LeCun joined the Adaptive Systems Research Department at AT&T Bell Laboratories, located in Holmdel, New Jersey—or in Murray Hill, New Jersey, according to Italian Wikipedia—under the leadership of Lawrence D. Jackel.^{[17,15,16,158,159]} During his time there, he became known as the founder of Convolutional Neural Networks (CNNs) by developing biologically inspired image recognition models, alongside the "Optimal Brain Damage" regularization method and Graph Transformer Networks (similar to conditional random fields).^{[17,15,16,158,159,160]} He applied these methods to handwriting recognition and optical character recognition (OCR), helping create a bank check recognition system widely deployed by NCR and other companies that read over 10% of all checks in the United States in the late 1990s and early 2000s.^{[17,15,16,158,159,160]} In 1990, he was elected a Fellow of the Association for the Advancement of Artificial Intelligence (AAAI). In 1996, he was appointed head of the Image Processing Research Department at AT&T Labs-Research, part of Lawrence Rabiner's Speech and Image Processing Research Lab, where he collaborated with Léon Bottou, Patrick Haffner, and Vladimir Vapnik, working primarily on DjVu image compression technology used by platforms such as the Internet Archive.^[18]

After a short stint as a fellow at the NEC Research Institute (now NEC-Labs America) in Princeton, New Jersey, LeCun moved to New York University (NYU) in 2003.^{[19,66,161,20]} At NYU, he served as Silver Professor of Computer Science and Neural Science at the Courant Institute of Mathematical Sciences and the Center for Neural Science, as well as a professor at the Tandon School of Engineering.^{[19,20,66,161]} His research at NYU focused on energy-based models for supervised and unsupervised learning, feature learning for object recognition in computer vision, mobile robotics, and algorithmic neuroscience.^{[21,22,23,20,51,19,66,161]} In 2012, he became the founding director of the NYU Center for Data Science (NYU-CDS).^{[24,26,162,51,233]} On December 9, 2013, LeCun was appointed the first director of Meta AI Research (formerly Facebook AI Research) in New York City, stepping down from directing NYU-CDS in early 2014 to maintain a part-time role at NYU, and later becoming Chief AI Scientist at FAIR in December 2017.^{[25,26,222,235]} In 2013, LeCun and Yoshua Bengio co-founded the International Conference on Learning Representations (ICLR), which adopted an open post-publication review process he had advocated on his website. Additionally, he chaired and organized the annual "Learning Workshop" in Snowbird, Utah from 1986 to 2012, serves on the Scientific Advisory Board of UCLA's Institute for Pure and Applied Mathematics, co-directs CIFAR's Learning in Machines & Brains program, and has advised companies including MuseAmi, KXEN Inc., and Vidient Systems.^{[27,28,52,234]} In 2016, he served as a visiting professor for the "Chaire Annuelle Informatique et Sciences Numériques" at the Collège de France in Paris, where his inaugural lecture was a major event in the city's intellectual life.^[29]

LeCun has been recognized with numerous honors, including election as a Fellow of the Association for Computing Machinery (2023) and an honorary doctorate of engineering from the Hong Kong University of Science and Technology (2023).^{[163,164,236]} He has expressed critical views on large data-driven AI models like ChatGPT, stating that "machine learning is crap", and instead advocates for the development of "World Models".^[57,58] In late 2025, LeCun announced his departure from Meta to establish an AI startup in Paris called AMI Labs (Advanced Machine Intelligence) alongside Nick Clegg, focusing on World Models.^[58,59,237] Over his career, his awards include the IEEE Neural Network Pioneer Award (2014), the PAMI Distinguished Researcher Award (2015), the IRI Medal (2018), the Harold Pender Award (2018), the ACM A.M. Turing Award (2018, awarded jointly with Bengio and Hinton), the Morris Loeb Lecture (2019), the Princess of Asturias Award (2022), the Global Swiss AI Award 2023, selection as Gibbs Lecturer 2025, and the Queen Elizabeth Prize for Engineering (2025), alongside election to the National Academy of Engineering (2017) and the National Academy of Sciences (2021).^{[13,60,61,62,63,6]}

LeCun's career has been spent primarily at Bell Labs, New York University, and Meta Platforms, Inc. In 1988, he began working at AT&T Bell Labs on optical character recognition software used for handling handwritten checks and helped develop convolutional neural networks (CNNs). In 2003, he became a professor and researcher at New York University, where he serves as Silver Professor at the Courant Institute of Mathematical Sciences and, in 2023, was appointed as the inaugural Jacob T. Schwartz Professor of Computer Science.^[50] Under his supervision, many doctoral students—including Wojciech Zaremba—prepared and defended their doctorates, making significant contributions to mathematics and computer science.^[198] In 2013, LeCun was hired as Director of Artificial Intelligence at Facebook (later Meta), where he served as Vice President and Chief AI Scientist and co-founded and led Meta AI.^[50] At Facebook, he worked on developing intelligent personal assistants capable of autonomously reading, translating text, and making reservations.^[127] In 2025, he announced his departure from Meta—following similar moves by prominent scientists such as Geoffrey Hinton and Fei-Fei Li—officially leaving in December 2025 to launch a new startup in Paris named AMI Labs (or Advanced Machine Intelligence).^[155,156] Upon its creation, the startup raised 3 billion euros, making it a triceratops, as a startup raising 1 billion euros is called a unicorn.^[48] Through AMI Labs in 2026, LeCun aims to foster a third AI revolution focused on understanding the physical world through world models, with applications in industry, robotics, and autonomous vehicles.

Recognized as a pioneer in artificial vision, artificial neural networks, and image recognition, LeCun has published over 130 papers and articles in these fields.^[123,168] He has stated that his research was notably inspired by Jean Piaget and constructivist theory.^[124] While his role in popularizing deep learning is undisputed, some scholars have criticized him for ignoring or glossing over the work of predecessors in his publications.^[126] Among his fundamental contributions, LeCun worked on developing and refining backpropagation, an essential algorithm for training neural networks in machine learning. His most notable contribution lies in the development of convolutional neural networks (CNNs), which are highly effective for processing visual data such as photos and images.^[99,125,133] CNNs have proven essential for computer vision tasks, including image recognition, document recognition, object detection, and semantic segmentation.^[99] In the 1990s, CNN technology was applied commercially by Crédit Mutuel de Bretagne for automated optical check reading.^[125,133] In addition to CNNs, LeCun participated in developing DjVu image compression technology and worked on recurrent neural networks (RNNs) for processing sequential data in natural language processing and industrial sensor applications.

In March 2019, LeCun received the 2018 Turing Award jointly with Yoshua Bengio and Geoffrey Hinton.^[3] In 2013, he and Yoshua Bengio co-founded the International Conference on Learning Representations (ICLR), an annual machine learning conference that adopted an open review process.^[78] LeCun and Bengio also co-founded and co-directed the Learning in Machines & Brains research program (formerly Neural Computation & Adaptive Perception), funded by CIFAR, which connects multidisciplinary international scientists to develop computers that can think more like humans.^[29] He co-organized CIFAR-funded summer workshops at the University of California, Los Angeles (UCLA) dedicated to machine learning, AI, and computational neuroscience, and serves on the Scientific Advisory Board of UCLA's Institute for Pure and Applied Mathematics.^{[200,199,219,29]} In 2016, as a visiting professor at the Collège de France in Paris, he delivered the inaugural lecture Deep learning, une révolution en Intelligence artificielle for the annual Computer and Digital Sciences course.^[201,202,85] LeCun serves as a scientific advisor to the French AI research group Kyutai, funded by Xavier Niel, Rodolphe Saadé, Eric Schmidt, and others, and was a co-founder and former board member of organizations researching the social consequences of artificial intelligence.^{[112,197,230,27]}

In 2015, LeCun estimated that advancing artificial intelligence requires integrating deep learning into reasoning and planning systems, unifying supervised and unsupervised learning under a single brain-inspired rule without pre-programmed cognitive structures.^[124] He asserts that artificial intelligence is not directly comparable to a biological human brain, as their characteristics differ fundamentally.^[126,66] Since 2019, he has devoted most of his research to self-supervised learning.^{[66,132,121,169]} Starting in 2022 at Meta, LeCun began developing the concept of open-source objective-driven AI using the I-JEPA (Joint Embedding Predictive Architecture) prediction model.^{[127,128,121,169,122]} By observing the world directly like young children, I-JEPA constructs abstract internal representations of essential elements to simulate and predict future events without being distracted by superficial details such as color, shape, or position.^[127,128] This internal world model allows AI to learn, plan, and perform complex tasks faster and adapt flexibly to context changes, offering significant applications for autonomous driving.^[129,132] In October 2024 at the University of Geneva, LeCun listed key challenges for AI to reach human-level intelligence, including massive training of world models, development of planning algorithms, and improvements to JEPA architectures with latent variables.^[131] He advocated refining the mathematical foundations of energy-based learning, replacing probabilistic models with energy-based models, abandoning generative models for perception-and-action architectures, moving away from contrastive methods, and limiting reinforcement learning strictly to adjusting world models.^[134,133]

LeCun has consistently denied alarmist narratives regarding superintelligent AI threatening humanity, viewing them as fanciful science fiction. He has openly criticized figures like Sam Altman, Demis Hassabis, and Dario Amodei for deliberately maintaining a climate of fear over AI catastrophe scenarios, insisting that AI systems are merely tools without consciousness, willpower, or understanding of the world.^[139] Opposed to regulation that he considers anachronistic, LeCun defended inclusive and transparent AI governance before the UN Security Council in December 2024, warning that doomsday rhetoric risks slowing innovation for the benefit of a few dominant actors or states.^[134] Instead, LeCun advocates for an open-source AI infrastructure similar to Linux—promoted by governments to ensure reliability—and continues to push for open-source foundation models to promote international cooperation encompassing all human languages and cultures.^[130,128] He foresees that everyone will soon have access to a domain-specific assistant smarter than themselves.

In 1988, Yann LeCun joined the Adaptive Systems Research Department at AT&T Bell Laboratories in Holmdel, New Jersey, United States, which was headed by Lawrence D. Jackel.^{[72,69,68,187,188,94]} At the lab, LeCun developed several new machine learning methods, including a biologically inspired model of image recognition called convolutional neural networks (LeNet), the "Optimal Brain Damage" regularization technique, and Graph Transformer Networks (a method similar to a conditional random field).^{[72,69,68,17,187,188,94]} He applied these techniques to handwriting recognition, optical character recognition (OCR), and document recognition.^{[72,69,68,95,17,187,188,94]}

The bank check recognition system that LeCun helped develop was widely deployed and commercialized by NCR and other companies.^[73,70,227] Starting in June 1996, NCR deployed large-scale check-reading machines at banks based on the results of their research.^[227] In the late 1990s and early 2000s, this system was evaluated to have processed over 10% of all checks in the United States.

In 1996, LeCun joined AT&T Labs-Research as head of the Image Processing Research Department, which was part of Lawrence Rabiner's Speech and Image Processing Research Lab.^[74] During this time, he worked primarily on DjVu image compression technology, a format designed for the efficient distribution of scanned documents across various websites, notably used by the Internet Archive to provide access to digitized texts.^{[74,18,185,10,186,75,228]} His major collaborators at AT&T included Léon Bottou and Vladimir Vapnik.

After a brief tenure as a research fellow at the NEC Research Institute (now NEC-Labs America) in Princeton, New Jersey, Yann LeCun joined New York University in 2003.^[75,76,19] At NYU, he taught computer science and held professorships at the Courant Institute of Mathematical Sciences, the Center for Neural Science, and the Tandon School of Engineering.^[75,76,19] While Polish sources state that he began his tenure at NYU in 2003 as the Jacob T. Schwartz Professor, other sources state he was named as the inaugural Jacob T. Schwartz Chaired Professor of Computer Science and Neural Science at the Courant Institute in 2023.^{[80,77,53,64]} He currently continues to work at New York University.

At NYU, his research has primarily focused on energy-based models for supervised and unsupervised learning, feature learning for object recognition in computer vision, machine learning, and mobile robotics.^{[27,20,51,21,22,189,191,66,192,25,190]} In 2012, LeCun became the founding director of the NYU Center for Data Science.^{[28,24,25,79,77]} On December 9, 2013, he became the first director and Chief AI Scientist of Meta AI Research (formerly Facebook AI Research) in New York City, subsequently stepping down from his directorship at the NYU Center for Data Science in early 2014.^{[29,64,66,25,24,26,77,193,194,196,80]}

In 2013, LeCun and Yoshua Bengio co-founded the International Conference on Learning Representations (ICLR), which adopted a post-publication open review process that LeCun had previously advocated on his website. Between 1986 and 2012, he organized and chaired the annual Learning Workshop held in Snowbird, Utah. He serves on the Scientific Advisory Board of the Institute for Pure and Applied Mathematics at UCLA and co-directs the Learning in Machines & Brains research program at CIFAR in Canada.^[78,28,27] In 2016, he was a visiting chair professor at the Collège de France in Paris for the annual chair in Computer Science and Digital Science, delivering the inaugural lecture.^[79,29] Additionally, LeCun serves as a scientific advisor to the French research organization Kyutai, which is funded by Xavier Niel, Rodolphe Saadé, Eric Schmidt, and others.^[53,77,78]

Following developments in generative artificial intelligence driven by transformer architectures and large language models, LeCun advocated that language models are not the proper path to artificial general intelligence because an architecture incorporating real-world models and rules is necessary. He has lectured and written extensively on the topic, releasing an initial implementation of his world-model architecture in July 2023. In November 2025, LeCun announced his intention to leave Meta to establish a startup focused on world models, and in March 2026, it was reported that the startup, AMI, raised over one billion dollars with LeCun as chairman and Alexandre LeBrun leading the company.^[66]

LeCun served as vice president and chief scientist of artificial intelligence at Meta, leading research in artificial intelligence and deep learning. On 19 November 2025, LeCun confirmed that he would be leaving Meta after ten years to found his own company focused on world-model architectures and human-like artificial intelligence he calls superintelligence.^[66]

The company he founded, Advanced Machine Intelligence Labs (or AMI Labs), is run by CEO Alex LeBrun, with LeCun serving as Executive Chair.^[83] This venture is focused on building AI "world models": systems that learn to understand the physical world's structure and dynamics rather than just predict text like large language models.^[84,81,229] In March 2026, AMI announced it had raised $1.03 billion in funding at a $3.5 billion pre-money valuation. The funding round was co-led by investors including Cathay Innovation, Greycroft, Hiro Capital, HV Capital and Bezos Expeditions.^{[30,85,82,28]}

On June 14, 2023, the newspaper Les Échos transcribed a debate between Yann LeCun and researcher Yoshua Bengio, an artificial intelligence specialist and founder of Mila.^[135] Regarding the risks affecting democracies, including the danger of opinion manipulation that worries Bengio, LeCun asserts that artificial intelligence will not amplify these problems, but will instead provide solutions to them.^[137,136]

His view that artificial intelligence cannot pose existential risks, mainly based on the idea that creators can explicitly build the desires of these systems, has been criticized on numerous occasions. For instance, in the book If Anyone Builds It, Everyone Dies, the authors explain that because current AI systems are grown rather than built, humans are incapable of understanding the inner workings of models resulting from the learning process, predicting their behavior, or choosing their goals and motivations.

Yann LeCun is a member of the US National Academy of Sciences and National Academy of Engineering (since 2017), as well as the French Académie des Sciences (since 2021).^{[30,170,216,218,217,35,231]} He has received honorary doctorates from the Instituto Politécnico Nacional (IPN) in Mexico City (2016), EPFL in Lausanne (October 2018), Université Côte d'Azur in Nice (2021), Università di Siena (2023), and the Hong Kong University of Science and Technology (2023).^{[34,31,32,33,35,86,41,37,232]}

He received the IEEE Neural Network Pioneer Award in 2014, the PAMI Distinguished Researcher Award in 2015, and has received several IEEE awards in the fields of neural networks and computer vision.^[36,87,88] In 2018, LeCun was awarded the IRI Medal, established by the Industrial Research Institute (IRI), and the Harold Pender Award given by the University of Pennsylvania in September of that year.^{[37,40,41,88,209,36]} Considered the greatest recognition of his life, LeCun won the 2018 Turing Award in March 2019, sharing it with Yoshua Bengio and Geoffrey Hinton for their groundbreaking contributions to deep learning.^{[36,34,90,39,157,89]} In September 2019, he received the Golden Plate Award from the American Academy of Achievement.^[36]

In 2017, LeCun declined an invitation to give lectures at King Abdullah University of Science and Technology in Saudi Arabia because he believed that, given his atheism, he would be considered a terrorist in the country.^[39,40] In 2022, he received the Princess of Asturias Award in the category of Technical and Scientific Research alongside Yoshua Bengio, Geoffrey Hinton, and Demis Hassabis.^{[38,91,101,6]} In 2023, the President of France made him a Chevalier (Knight) of the French Legion of Honour.^{[92,89,90,93]} During the World Economic Forum (WEF) 2024 in Davos, he received the Global Swiss AI Award 2023.^[93,91] That same year, he received the Grand Prize of the VinFuture Prize alongside Yoshua Bengio, Jensen Huang, Geoffrey Hinton, and Fei-Fei Li for their groundbreaking contributions to neural networks and deep learning algorithms.^[3,67] In 2025, he was awarded the Queen Elizabeth Prize for Engineering jointly with Yoshua Bengio, Bill Dally, Geoffrey E. Hinton, John Hopfield, Jensen Huang, and Fei-Fei Li.

Related categories for this profile include Turing Award laureates, Knights of the Legion of Honour, and members of the French Academy of Sciences. Additionally, biographical classifications include people born in Paris, French engineers, as well as American computer scientists, programmers, and scientists.