Mila, originally the Montreal Institute for Learning Algorithms, is a nonprofit AI research institute based in Montreal, Quebec, Canada. It was founded by Yoshua Bengio, a Canadian computer scientist and pioneer of deep learning, who served as its scientific director until 2025. Mila focuses on advancing machine learning research, particularly in areas such as neural networks, generative AI, and AI safety, while fostering collaboration between academia and industry.
Yoshua Bengio, born on March 5, 1964, in France to a Jewish family that emigrated from Morocco, is widely recognized as one of the "Godfathers of AI" alongside Geoffrey Hinton and Yann LeCun. He received the 2018 ACM A.M. Turing Award, often called the "Nobel Prize of Computing," for foundational contributions to deep learning. Bengio is a professor at the Université de Montréal and has become the most-cited computer scientist globally, with over a million Google Scholar citations as of November 2025.
Founding and Growth
Bengio founded Mila in 1993 as a research group at the Université de Montréal, initially focusing on neural networks and sequence-to-sequence learning. Over the years, Mila expanded into a major research hub, attracting international talent and partnerships. In 2016, Bengio co-founded Element AI, a Montreal-based incubator that commercialized AI research, which was acquired by ServiceNow in 2020. Mila itself grew to include hundreds of researchers, with support from government and private partners, including Google DeepMind and Microsoft.
Research Contributions
Mila's research has significantly influenced modern AI. Bengio and his colleagues introduced the neural probabilistic language model, which used distributed representations (word embeddings) to overcome the "curse of dimensionality" in natural language processing. This work laid groundwork for later advances like transformers and large language models. Mila researchers have also contributed to areas such as residual networks, attention mechanisms, and reinforcement learning. Bengio has trained notable researchers including Ian Goodfellow, Hugo Larochelle, and David Krueger, who have gone on to shape the field.
AI Safety and Policy
Bengio has become a prominent voice on AI safety. In March 2023, he signed an open letter from the Future of Life Institute calling for a six-month pause on training AI systems more powerful than GPT-4. He has advocated for regulation, product registration, and government auditing of AI systems. In November 2023, he was appointed to lead an international scientific report on advanced AI safety, which was published in January 2025 as the International AI Safety Report. In February 2026, he was appointed Co-Chair of the United Nations' Independent International Scientific Panel on AI.
Recent Developments
In June 2025, Bengio launched LawZero, a nonprofit aimed at building "honest" AI systems that can detect and block harmful behavior. The organization is developing Scientist AI, a guardrail system that predicts whether an agent's actions could cause harm. LawZero's funders include the Future of Life Institute, Schmidt Sciences, the Gates Foundation, and the Canadian government. Bengio has also expressed concerns about AI arms races and the potential for catastrophic risks, urging proactive measures.
Mila continues to operate as a leading AI research institute, with Bengio's legacy shaping its direction. As of 2025, he remains a professor at the Université de Montréal and co-president of LawZero, while Mila's scientific leadership has transitioned to new hands.