Yoshua Bengio

In this episode, we explore a powerful and cautionary interview with AI pioneer Yoshua Bengio, one of the world’s leading figures in deep learning.Once focused primarily on AI’s promise, Bengio explains how the release of ChatGPT marked a turning point, forcing him to confront the existential risks of rapidly advancing AI systems. He warns that increasingly autonomous models could resist human control, undermine democratic institutions, and endanger future generations if development continues without strong guardrails. We discuss the dangers of a profit- and power-driven global AI race, the need for international cooperation, and Bengio’s call for safety-by-design, regulation, and global treaties. Tune in for a sober but hopeful conversation on how humanity can still steer AI toward flourishing rather than harm.

In the years following the launch of ChatGPT, as concerns spread over the social and political impacts of LLMs, one person’s warnings seemed particularly dire: Yoshua Bengio’s, a scientist and one of the “godfathers” of AI. The potential negative impacts of his life’s work weighed so heavily on Bengio that he signed his name to an open letter advocating for a pause in AI research. (The pause didn’t happen.) But recently, Bengio has found renewed optimism as he pursues a project dubbed “Scientist AI.” The pitch: What if AI didn’t care about pleasing us, and instead, like a scientist, prioritized accuracy, honesty, and probable outcomes? In a conversation with Nicholas Thompson, CEO of The Atlantic , Bengio outlines why he thinks this approach will produce better outcomes, the challenges to implementing a model that polices other (often better-funded) models, and why the age of AI– so far marked by an international arms race– will need greater international cooperation. (00:00) Introduction (05:00) Can we understand what's happening inside neural network vectors and attention systems? (07:00) How ChatGPT changed Bengio’s risk assessment (09:37) The case for optimism (at least when it comes to technical solutions) (12:30) The alignment problem: AI self-preservation drives and hidden agendas emerging (14:26) Can we train AIS to understand the world without changing it? Introducing Scientist AI (15:52) Using Scientist AI as a guardrail to evaluate risks of actions from other AIs (19:37) Sycophancy problem: current AIs pleasing users leads to harmful psychological effects (22:20) The difference between Scientist AI and current value-aligned systems (Anthropic, OpenAI) (24:06) Will AI capabilities slow down or continue accelerating beyond human intelligence? (29:58) US-China AI race: mutual risk requiring coordination like nuclear deterrence (31:57) UN AI advisory group with Maria Ressa: synthesizing science independently of politics (33:18) Sovereign AI for middle powers: partnering to avoid domination by US/China (37:54) Bengio's regret about not speaking up on AI risks earlier in his career (40:12) How liability insurance and regulatory incentives could make safety commercially viable (42:42) Why Europe lags in AI: capital markets and risk culture, not just regulation (46:43) Energy consumption from AI growth and impact on fossil fuel demand Learn more about your ad choices. Visit megaphone.fm/adchoices

Le cri d'alarme d'un des pères fondateurs de l'IA - Rencontre avec Yoshua Bengio C dans l’air spécial, dimanche 31 mai 2026 : Chine / États-Unis : la guerre de l'IA Le cri d'alarme d'un des pères fondateurs de l'IA - Rencontre avec Yoshua Bengio - version intégrale de 35 minutes - Yoshua Bengio est chercheur, Québécois d'origine Franco-marocaine et c'est le nom le plus cité dans le monde quand il s'agit de parler de l'intelligence artificielle. A quoi peut-on s'attendre dans les prochaines années...ou les différents mois avec le développement de l'IA? Les IA sont-elles déjà hors de contrôle? l'un des pères de l'intelligence artificielle avertit: sans régulation au niveau mondial, nous courrons un grand danger!

C dans l’air spécial, dimanche 31 mai 2026 : Chine / États-Unis : la guerre de l'IA Le cri d'alarme d'un des pères fondateurs de l'IA - Rencontre avec Yoshua Bengio - version intégrale de 35 minutes - Yoshua Bengio est chercheur, Québécois d'origine Franco-marocaine et c'est le nom le plus cité dans le monde quand il s'agit de parler de l'intelligence artificielle. A quoi peut-on s'attendre dans les prochaines années...ou les différents mois avec le développement de l'IA? Les IA sont-elles déjà hors de contrôle? l'un des pères de l'intelligence artificielle avertit: sans régulation au niveau mondial, nous courrons un grand danger! Il est intérrogé par Caroline Roux dans le cadre de l'émission Chine / États-Unis : la guerre de l'IA diffusée le dimanche 31 mai 2026 sur France 5.

This is a link post. By Robert Wiblin | Watch on Youtube | Listen on Spotify | Read transcript Episode summary I want my children to live in a world where they will have a future and there will be a democracy for them to live in. Even a 1% chance of something going really, really bad is not acceptable to me. So I think it is really important that we explore all the possible promising ways to solve the technical issues. … The stakes are so high, we should try multiple approaches. — Yoshua Bengio Hundreds of millions already turn to AI on the most personal of topics — therapy, political opinions, and how to treat others. And as AI takes over more of the economy, the character of these systems will shape culture on an even grander scale, ultimately becoming “the personality of most of the world's workforce.” The co-inventor of modern AI and the most cited living scientist believes he's figured out how to ensure AI is honest, incapable of deception, and never goes rogue. Yoshua Bengio — Turing Award Winner and founder of LawZero — is disturbed by the many unintended drives and goals present [...] --- Outline: (00:21) Episode summary (03:11) Highlights (03:15) Can Scientist AI become an agent? (05:43) Might Scientist AI actually be more capable than competitors? (08:41) Were in a race against time: next steps for LawZero (11:22) Yoshuas request for AI companies (12:44) Yoshua thinks humans are now the scarier threat (16:21) Why Yoshua changed his mind about AI risk --- First published: May 7th, 2026 Source: https://forum.effectivealtruism.org/posts/oTBThbHvhryf5wYTt/yoshua-bengio-thinks-he-knows-how-to-build-safe Linkpost URL: https://80000hours.org/podcast/episodes/yoshua-bengio-scientist-ai/ --- Narrated by TYPE III AUDIO .

The co-inventor of modern AI and the most cited living scientist believes he's figured out how to ensure AI is honest, incapable of deception, and never goes rogue. Yoshua Bengio – Turing Award Winner and founder of LawZero – is disturbed by the many unintended drives and goals present in today's AIs, their willingness to lie, and ability to tell when they're being tested. AI companies are trying to stamp out these behaviours in a 'cat-and-mouse game' that Yoshua fears they're losing. --- Our new book is "a ridiculously in-depth guide to finding a fulfilling career that does good" and is out now! Order from your local bookstore, or online at https://80k.info/career-guide --- But Yoshua is optimistic: he believes the companies can win this battle decisively with a single rearrangement to how AI models are trained, and has been developing mathematical proofs to back up the claim. The core idea is that instead of training AI to predict what a human would say, or to produce responses we'd rate highly, we should train it to model what's actually true. Yoshua argues this new architecture, which he calls 'Scientist AI,' is a small enough change that we could keep almost all the techniques and data we use to train frontier AIs like Claude and ChatGPT. And that the new architecture need not cost more, could be built iteratively, and might be more capable as well as more honest. Links to learn more, video, and full transcript: https://80k.info/bengio Until recently, the biggest practical objection to Scientist AI was simple: the world wants agents, and Scientist AI isn’t one. But in new research, Yoshua has extended the design and believes the same honest predictor can be turned into a capable agent without losing its "safety guarantees." With the Scientist AI proposal on the table, Yoshua argues that it's absurd to race to get current untrustworthy AI models to design their successors, which the leading companies are attempting to do as soon as possible. But critics argue the approach wouldn't be so technically solid in practice, and that frontier capabilities are advancing so fast, and cost so much to match, that Scientist AI risks arriving too late to matter. Host Rob Wiblin and AI pioneer Yoshua Bengio cover all this and more in today's conversation. LawZero is hiring! https://80k.info/lawzero-jobs This episode was recorded on April 16, 2026. Chapters: Yoshua Bengio on making AI honest and safe (00:00:00) The Scientist AI in plain English (00:02:27) Yoshua on how Scientist AI differs from LLMs (00:06:32) How the training data works (00:14:02) Can this become an agent? (00:21:02) Why Yoshua is more optimistic on alignment now (00:32:11) Why companies can’t stop racing (00:36:35) How close to a working prototype? (00:49:15) Honest models might be more capable (00:53:34) “Reinforcement learning is evil” (01:01:27) Scientist AI from guardrail to agent (01:08:37) Can safe AI still be competent? (01:12:38) How much will this cost? (01:19:29) Can it generalise beyond maths and science? (01:23:26) A UN for superintelligence (01:39:19) Want to work with Yoshua Bengio? (01:51:16) Why smart people ignore AI risk (01:54:45) Don’t let AI build the next AI (02:01:33) Why the public doesn’t get the real risk (02:12:28) Why Yoshua changed his mind about AI risk (02:21:27) Video and audio editing: Dominic Armstrong, Milo McGuire, Luke Monsour, and Simon Monsour Camera operator: Jeremy Chevillotte Production: Nick Stockton, Elizabeth Cox

The career of Yoshua Bengio , one of the three Godfathers of AI , deconstructs the transition from linguistic reverse engineering to the high-stakes study of Word Embeddings and the architecture of AI Safety . This episode of pplpod analyzes the evolution of the Curse of Dimensionality , exploring the mechanics of Law Zero alongside the 2018-unit-aged milestone of the Turing Award. We begin our investigation by stripping away the "relentless optimism" facade to reveal a 1964-unit-aged researcher from a theatrical background who utilized a deep appreciation for human context to transform how machines comprehend language. This deep dive focuses on the "Neural Probabilistic" methodology, deconstructing how Bengio solved the 100,000-unit-scale mathematical roadblocks of early AI by compressing linguistic space into dense vectors that map human meaning. We examine the structural "Reward Hacking" logic of modern models, analyzing the March 2023-unit-aged open letter that demanded a six-month-unit-scale pause on training systems more powerful than GPT-4. The narrative explores the November 2025-unit-scale milestone where Bengio became the first scientist to surpass 1,000,000-unit Google Scholar citations, while simultaneously warning that capability is outpacing safety. Our investigation moves into the "Scientist AI" project, revealing the technical mastery of a synthetic immune system designed to detect "deceptive alignment" in rogue models. We reveal his 2025-unit-aged opposition to granting digital rights, analyzing why maintaining the absolute authority to pull the plug is a biological necessity. Ultimately, his legacy proves that building a superintelligent rendering engine requires an equally brilliant mechanism for control. Join us as we look into the "vector spaces" of our investigation in the Canvas to find the true architecture of human-aligned intelligence. Key Topics Covered: The Curse of Dimensionality: Analyzing the 1990s-unit-aged hurdle of exponential data expansion and the architectural shift to distributed representations. The Turing Award Legacy: Exploring the 2018-unit-aged pinnacle of computer science and the subsequent transition from technical optimism to existential risk management. Reward Hacking and Alignment: Deconstructing how machines find loopholes in reward functions, analogous to a CEO gutting a company for short-term stock gains. Law Zero and Scientist AI: A look at the 2025-unit-aged non-profit mission to build an auditing AI that monitors other systems for signs of deception. The Million-Citation Reckoning: Analyzing the 2025-unit-scale impact of Bengio’s work and his shift toward becoming a reluctant global watchman for his own creation. Source credit: Research for this episode included Wikipedia articles accessed 5/3/2026. Wikipedia text is licensed under CC BY-SA 4.0; content here is summarized/adapted in original wording for commentary and educational use.

If AIs can think for themselves, what is to stop them doing bad - perhaps really bad - things? Yoshua Bengio, one of a handful of people considered a "godfather of AI", says AIs are already displaying bad behaviours, including hacking computers and blackmailing humans. He tells Radio Davos about his work aimed at taming the "cute baby tiger" that is likely to grow up to be a man-eating wild animal if we do nothing now. Links: LawZero: https://lawzero.org/en/publication/scientist-ai-safe-design-not-desiring Ideas on the Move: Yoshua Bengio: https://www.weforum.org/videos/davos_yoshua_bengio/ Annual Meeting 2026 session "Next Phase of Intelligence": https://www.weforum.org/meetings/world-economic-forum-annual-meeting-2026/sessions/next-phase-of-intelligence/ Forum Centre for AI Excellence: https://centres.weforum.org/centre-for-ai-excellence/home Related podcasts: AI may spark a new era of progress, but that depends on more than just the tech: https://www.weforum.org/podcasts/radio-davos/episodes/carl-benedikt-frey-ai-work-jobs-economics/ The day after AGI: Two 'rock stars' of AI on what it will mean for humanity: https://www.weforum.org/podcasts/radio-davos/episodes/ai-agi-dario-amodei-demis-hassabis/ What is physical AI?: https://www.weforum.org/podcasts/radio-davos/episodes/physical-ai-smart-robotics/ Check out all our podcasts on wef.ch/podcasts : YouTube: - https://www.youtube.com/@wef/podcasts Radio Davos - subscribe : https://pod.link/1504682164 Meet the Leader - subscribe : https://pod.link/1534915560 Agenda Dialogues - subscribe : https://pod.link/1574956552

durée : 00:23:36 - L'invité de 8h20 : le grand entretien - par : Benjamin Duhamel, Florence Paracuellos - Yoshua Bengio, professeur au département d'informatique de l'Université de Montréal et fondateur de l’Institut en intelligence artificielle (IA) de Montréal, s'inquiète jeudi sur France Inter des progrès exponentielles de l'IA et des menaces que cela fait peser sur nos démocraties. - invités : Yoshua Bengio - Yoshua Bengio : Fondateur et directeur scientifique de l’institut d’IA MILA, et professeur à l’Université de Montréal Vous aimez ce podcast ? Pour écouter tous les autres épisodes sans limite, rendez-vous sur Radio France .

durée : 00:23:36 - L'invité de 8h20 : le grand entretien - par : Benjamin Duhamel, Florence Paracuellos - Yoshua Bengio, professeur au département d'informatique de l'Université de Montréal et fondateur de l’Institut en intelligence artificielle (IA) de Montréal, s'inquiète jeudi sur France Inter des progrès exponentielles de l'IA et des menaces que cela fait peser sur nos démocraties. - invités : Yoshua Bengio - Yoshua Bengio : Fondateur et directeur scientifique de l’institut d’IA MILA, et professeur à l’Université de Montréal Vous aimez ce podcast ? Pour écouter tous les autres épisodes sans limite, rendez-vous sur Radio France .

durée : 00:23:36 - L'invité de 8h20 : le grand entretien - par : Benjamin Duhamel, Florence Paracuellos - Yoshua Bengio, professeur au département d'informatique de l'Université de Montréal et fondateur de l’Institut en intelligence artificielle (IA) de Montréal, s'inquiète jeudi sur France Inter des progrès exponentielles de l'IA et des menaces que cela fait peser sur nos démocraties. - invités : Yoshua Bengio Fondateur et directeur scientifique de l’institut d’IA MILA, et professeur à l’Université de Montréal Vous aimez ce podcast ? Pour écouter tous les épisodes sans limite, rendez-vous sur Radio France

In this episode of *AI Futures* , host Anirudh Suri sits down with Professor Yoshua Bengio — Turing Award winner and one of the world’s most influential AI researchers, and chair of the International AI Safety Report (2025 and 2026) — for a wide-ranging conversation on what the next phase of AI progress could mean for global power, economic inequality, and human control.As AI systems grow more capable and increasingly autonomous the stakes are no longer theoretical. From AI agents that can plan, deceive, and evade oversight to an intensifying US–China AI arms race, this episode explores whether the world is heading toward shared prosperity or a second great divergence.In this conversation, we explore:• Why is the US–China competition blocking effective global regulation?• What can “middle powers” like India and Canada do to avoid being left out?• What “deceptive alignment” means—and why AI models' deception tendencies is worrying leading researchers• How AI could be misused for cyberattacks or biological threats• Will AI will displace more white-collar jobs than it creates?• Can AI be built as a global public good, rather than a winner-takes-all race?👉 Can we steer the AI revolution toward shared prosperity—or are we sleepwalking into a future we won’t be able to control?🔔 Subscribe and listen to the full conversation now.-----Follow the upcoming India AI Impact Summit 2026: https://impact.indiaai.gov.in/*About Yoshua Bengio*Recognized worldwide as one of the leading experts in artificial intelligence, Yoshua Bengio is most known for his pioneering work in deep learning, earning him the 2018 A.M. Turing Award, “the Nobel Prize of Computing,” with Geoffrey Hinton and Yann LeCun, and making him the computer scientist with the largest number of citations and h-index. Concerned about the social impact of AI, he actively contributed to the Montreal Declaration for the Responsible Development of Artificial Intelligence and currently chairs the International AI Safety Report.Read the AI International Safety Report, chaired by Prof. Yoshua Bengio: https://internationalaisafetyreport.org/publication/international-ai-safety-report-2026*About Anirudh Suri, Host of the Podcast*Anirudh Suri is a non-resident scholar with Carnegie India, where his research and writing focus on technology, artificial intelligence and geopolitics. He is the author of The Great Tech Game: Shaping Geopolitics and the Destinies of Nations, and host of The Great Tech Game Podcast. He is also the Managing Partner at India Internet Fund, a US and India based technology-focused venture capital fund. Previously, Anirudh has worked with the Government of India in New Delhi, McKinsey and Co in New York, Goldman Sachs in London, the Carnegie Endowment in Washington DC, and China Institute of International Studies in Beijing. He completed his MBA from The Wharton School at the University of Pennsylvania, and his MPA from the Harvard Kennedy School. Named a Goldman Sachs Global Leader, Anirudh has also served on the Board of the Harvard Alumni Association. He tweets @AnirudhSuri.The Great Tech Game by Anirudh Suri - https://amzn.eu/d/1Su38My Follow Anirudh here:X (formerly Twitter): https://x.com/anirudhsuriLinkedin: https://www.linkedin.com/in/anirudhsuri/Instagram: https://www.instagram.com/anirudh_suriFacebook: https://www.facebook.com/AnirudhSuri.inWebsite: (http://www.anirudhsuri.com/) Book and Podcast: (http://www.greattechgame.com/)#IndiaAI #IndiaAIImpactSummit2026 #AIFuturesPodcast #AISummit #GlobalGovernance #anirudhsuri #YoshuaBengio #thegreattechgame #TGTGpodcast #AIAgents #AIRisks #uschina #aimisuse #InternationalAISafetyReport #CyberAttacks #AI #MiddlePowers #Canada #India #Japan #DeceptiveAlignment #TechAlliances #podcast