Yoshua Bengio of the University of Montreal has become the first researcher to exceed one million Google Scholar citations. The milestone underscores the rapid rise of machine learning — eight of the ten most-cited papers this century concern ML. Bengio’s top-cited work includes the 2014 "Generative Adversarial Nets" paper and the 2015 Nature review "Deep Learning". Experts praise the achievement while warning that raw citation totals are an imperfect and sometimes manipulable metric.
‘Godfather of AI’ Yoshua Bengio Surpasses One Million Google Scholar Citations
Yoshua Bengio of the University of Montreal has become the first researcher to exceed one million Google Scholar citations. The milestone underscores the rapid rise of machine learning — eight of the ten most-cited papers this century concern ML. Bengio’s top-cited work includes the 2014 "Generative Adversarial Nets" paper and the 2015 Nature review "Deep Learning". Experts praise the achievement while warning that raw citation totals are an imperfect and sometimes manipulable metric.
02:14 AM, 11/14/2025Science
Yoshua Bengio becomes the first researcher to pass one million Google Scholar citations
Computer scientist Yoshua Bengio, based at the University of Montreal, has become the first researcher whose work has been cited more than one million times on Google Scholar. The milestone spotlights both Bengio’s long-standing influence on machine learning and the explosive growth of the field in recent years.
Bengio is widely regarded as one of the field’s “godfathers” of artificial intelligence, alongside Geoffrey Hinton and Yann LeCun. The three shared the A. M. Turing Award in 2019 for foundational advances in neural networks that underpin much modern AI.
Key works and influence
Among Bengio’s most-cited papers are widely studied contributions such as the 2014 paper "Generative Adversarial Nets" (co‑authored with Ian Goodfellow and others), which has amassed over 105,000 Google Scholar citations, and the influential 2015 Nature review "Deep Learning", co-authored with Yann LeCun and Geoffrey Hinton. His work on attention mechanisms also appears among his most-cited items; attention later became a central innovation powering large language models and the chatbot wave that gained mainstream attention with ChatGPT in 2022.
Expert reactions and caveats
“AI is changing the world, and we’re just seeing the tip of the iceberg,”
Researchers have called the milestone remarkable. Kaiming He, a computer scientist who co-authored what a Nature analysis identified as the most-cited paper of the twenty-first century, notes the dominance of machine learning in recent high-impact work: eight of the ten most-cited papers this century were on machine learning.
At the same time, experts caution that raw citation counts are an imperfect metric. Alberto Martín Martín, an information scientist at the University of Granada, says citation totals are “crude metrics” that can be gamed and cautions institutions against using such rankings solely for marketing.
Why platforms disagree
Different bibliometric services use different coverage and counting rules. Platforms such as Web of Science, Scopus and OpenAlex typically report lower totals because they focus on curated, peer‑reviewed literature. Google Scholar counts a broader range of sources — including books, conference papers and preprints posted online — which usually yields higher citation numbers.
Bengio on metrics and tools
Bengio says he is an “avid user” of Google Scholar, which recently celebrated its twentieth anniversary. He credited the service with making literature discovery far less painstaking, while warning that researchers should not treat citation counts as an objective in themselves: “It should not become an objective for researchers to have more citations, because it leads into trying to optimize this rather than do good science and go after the truth.”
This article is reproduced with permission and was first published on November 12, 2025.