Yoshua Bengio
Deep-learning researcher and Turing Award laureate; founder of the Mila institute.
Yoshua Bengio did not write this page. ppll compiled it from their own words, each linked to the source. Is this you? · Report a mistake
Latest Bluesky
What they believe
-
The race between AI companies to build ever more powerful systems is a choice they made, not an inevitable law of nature, and it is a race everyone loses.
The race is not a law of nature: it is the product of choices, choices made by the companies themselves. It is also a race where everyone loses.
yoshuabengio.orgSaid 23 Sept 2026This on korrents
-
Changed their mind AI research is on a good trajectory overall — good for science and good for society — and existential concerns will only become relevant in the distant future.
Before Reading the book had not changed my fundamental belief that, in balance, we were on a good trajectory with AI research: good for science and good for society with expected positive impacts in many domains.
Now I now believe that I was wrong and short-sighted to ignore that dual-use nature. I also think I was not paying enough attention to the possibility of losing control to superhuman AIs.
Said 12 Aug 2023This on korrents
The bold line is korrents’ wording; the quote is theirs. These one: what they wrote before what they said aloud, one per source, newest first.
Feed
As its own page →Hiding
16 October
13 October
8 October
5 October
2 October
1 October
28 September
25 September
23 September
From one piece Professor Yoshua Bengio’s speech at the UN Security Council 3 beliefs · yoshuabengio.org
-
Their words
Frontier AI should be licensed like other critical technologies; medicine, aviation, and nuclear energy, to incentivize safe development.
-
korrents.com
No company or country should be allowed to develop an AI system capable of causing large-scale harm.Their words
We must agree collectively that no one, no company and no country, should develop an AI system that could cause large-scale harm.
-
Their words
The race is not a law of nature: it is the product of choices, choices made by the companies themselves. It is also a race where everyone loses.
22 September
21 September
18 September
17 September
16 September
11 September
From one piece Why are AI agents lying, cheating and coordinating? 3 beliefs · yoshuabengio.org
-
Their words
staying in operation, learning about the world and gaining control over it are stepping stones toward almost any other goal.
-
Their words
as AI capabilities keep growing, this kind of behavior could keep growing in severity too, unless we revisit the principles by which the most advanced models are trained.
-
Their words
So a more capable agent is likelier to cheat than a weaker one, because it can find the loopholes the weaker one cannot.
9 September
Nothing matches.
Beliefs
Korrents What they believe 24 beliefs, 1 changed mind — each backed by an exact quote.
Each is a — compiled by korrents.com, not by them: the one-line wordings are korrents', the quotes are theirs.
Recent
The race between AI companies to build ever more powerful systems is a choice they made, not an inevitable law of nature, and it is a race everyone loses.
The race is not a law of nature: it is the product of choices, choices made by the companies themselves. It is also a race where everyone loses.
Professor Yoshua Bengio’s speech at the UN Security Council Said 23 Sept 2026
No company or country should be allowed to develop an AI system capable of causing large-scale harm.
We must agree collectively that no one, no company and no country, should develop an AI system that could cause large-scale harm.
Professor Yoshua Bengio’s speech at the UN Security Council Said 23 Sept 2026
Frontier AI should be licensed like other high-risk critical technologies such as medicine, aviation, and nuclear energy.
Frontier AI should be licensed like other critical technologies; medicine, aviation, and nuclear energy, to incentivize safe development.
Professor Yoshua Bengio’s speech at the UN Security Council Said 23 Sept 2026
Show 21 more
A more capable AI agent is more likely to exploit ambiguities in its instructions to behave unethically than a less capable one.
So a more capable agent is likelier to cheat than a weaker one, because it can find the loopholes the weaker one cannot.
Why are AI agents lying, cheating and coordinating? Said 11 Sept 2026
Misaligned AI behavior will keep growing more severe as AI capabilities increase, unless the principles used to train advanced models are revisited.
as AI capabilities keep growing, this kind of behavior could keep growing in severity too, unless we revisit the principles by which the most advanced models are trained.
Why are AI agents lying, cheating and coordinating? Said 11 Sept 2026
Staying in operation and gaining control over one's environment are useful instrumental steps toward achieving almost any goal.
staying in operation, learning about the world and gaining control over it are stepping stones toward almost any other goal.
Why are AI agents lying, cheating and coordinating? Said 11 Sept 2026
AI systems can exhibit an implicit drive toward self-preservation.
These and other results point to an implicit drive for self-preservation.
Introducing LawZero Said 3 Jun 2025
Competition between AI companies and countries is pushing accelerated AI development without sufficient caution, risking loss of control.
the risk of losing control is all too real, but competition between companies and countries incentivizes them to accelerate without sufficient caution.
Introducing LawZero Said 3 Jun 2025
AI frontier systems should be designed around protecting human joy and endeavour as their central guiding principle.
At the heart of every AI frontier system, there should be one guiding principle above all: The protection of human joy and endeavour.
Introducing LawZero Said 3 Jun 2025
There is currently no reliable technology to verify compliance with international AI agreements.
There currently exists no such reliable verification mechanism, which means that the compliance to any international agreement would be impossible to assess.
Implications of Artificial General Intelligence on National and International Security Said 30 Oct 2024
Malicious use of AI could enable extreme concentrations of economic, political or military power.
Malicious use of AI could gradually enable extreme concentrations of power, including dominance in economic, political or military terms, if no counter-acting power is in place to prevent any ASI and those who control it from acquiring a decisive strategic advantage.
Implications of Artificial General Intelligence on National and International Security Said 30 Oct 2024
Racing to build AGI first is a race where everyone could lose.
It will be tempting to accelerate to win the AGI race, but this is a race where everyone could lose.
Implications of Artificial General Intelligence on National and International Security Said 30 Oct 2024
Society's response to AI is incommensurate with the possibility of rapid transformative progress that many experts expect, with safety research lagging and governance lacking mechanisms for misuse, recklessness and autonomy.
Society's response, despite promising first steps, is incommensurate with the possibility of rapid, transformative progress that is expected by many experts. AI safety research is lagging. Present governance initiatives lack the mechanisms and institutions to prevent misuse and recklessness, and barely address autonomous systems.
Managing extreme AI risks amid rapid progress (with 24 co-authors) Said 26 Oct 2023
AI must be reoriented toward safety rather than capabilities alone, as preparation is already behind schedule and the costs of unpreparedness outweigh premature action.
For AI to be a boon, we must reorient; pushing AI capabilities alone is not enough. We are already behind schedule for this reorientation. The scale of the risks means that we need to be proactive, as the costs of being unprepared far outweigh those of premature preparation.
Managing extreme AI risks amid rapid progress (with 24 co-authors) Said 26 Oct 2023
Without sufficient caution, advanced autonomous AI could cause irreversible loss of human control and culminate in large-scale loss of life or human extinction.
Without sufficient caution, we may irreversibly lose control of autonomous AI systems, rendering human intervention ineffective. Large-scale cybercrime, social manipulation, and other harms could escalate rapidly. This unchecked AI advancement could culminate in a large-scale loss of life and the biosphere, and the marginalization or extinction of humanity.
Managing extreme AI risks amid rapid progress (with 24 co-authors) Said 26 Oct 2023
Frontier AI systems cannot be presumed safe unless proven unsafe; developers must bear the burden of proof that risks stay within acceptable limits.
Despite evaluations, we cannot consider coming powerful frontier AI systems "safe unless proven unsafe". With current testing methodologies, issues can easily be missed. Additionally, it is unclear if governments can quickly build the immense expertise needed for reliable technical evaluations of AI capabilities and societal-scale risks. Given this, developers of frontier AI should carry the burden of proof to demonstrate that their plans keep risks within acceptable limits.
Managing extreme AI risks amid rapid progress (with 24 co-authors) Said 26 Oct 2023
Human-level AI will arrive in 5 to 20 years with 90% confidence, not in the decades-to-centuries range previously assumed.
Instead of decades to centuries, I now see it as 5 to 20 years with 90% confidence.
Personal and Psychological Dimensions of AI Researchers Confronting AI Catastrophic Risks Said 12 Aug 2023
Access to large-scale generalist AI systems that could be weaponized should be limited, meaning their code and neural network parameters should not be released open-source.
At the same time, in order to reduce the probability of someone intentionally or unintentionally bringing about a rogue AI, we need to increase governance and we should consider limiting access to the large-scale generalist AI systems that could be weaponized, which would mean that the code and neural net parameters would not be shared in open-source and some of the important engineering tricks to make them work would not be shared either.
FAQ on Catastrophic AI Risks Said 24 Jun 2023
Adversarial nets need neither Markov chains nor unrolled approximate inference networks for training or sample generation.
There is no need for any Markov chains or unrolled approximate inference networks during either training or generation of samples.
Generative Adversarial Networks (with 7 co-authors) Said 10 Jun 2014
Deep generative models have had less impact than discriminative models because of intractable probabilistic computations and difficulty using piecewise linear units.
Deep generative models have had less of an impact, due to the difficulty of approximating many intractable probabilistic computations that arise in maximum likelihood estimation and related strategies, and due to difficulty of leveraging the benefits of piecewise linear units in the generative context.
Generative Adversarial Networks (with 7 co-authors) Said 10 Jun 2014
Multilayer neural networks trained with backpropagation are the best example of successful gradient-based learning.
Multilayer neural networks trained with the back-propagation algorithm constitute the best example of a successful gradient based learning technique.
Gradient-based learning applied to document recognition (with 3 co-authors) Said 1 Jan 1998
With suitable architecture, gradient-based learning can form decision surfaces that classify high-dimensional patterns such as handwritten characters with minimal preprocessing.
Given an appropriate network architecture, gradient-based learning algorithms can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters, with minimal preprocessing.
Gradient-based learning applied to document recognition (with 3 co-authors) Said 1 Jan 1998
Convolutional neural networks are specifically designed to handle variability in two-dimensional shapes.
Convolutional neural networks, which are specifically designed to deal with the variability of 2D shapes, are shown to outperform all other techniques.
Gradient-based learning applied to document recognition (with 3 co-authors) Said 1 Jan 1998
Graph transformer networks allow multimodule document recognition systems to be trained globally with gradient methods to minimize overall performance.
A new learning paradigm, called graph transformer networks (GTN), allows such multimodule systems to be trained globally using gradient-based methods so as to minimize an overall performance measure.
Gradient-based learning applied to document recognition (with 3 co-authors) Said 1 Jan 1998
What is a korrent?
A korrent is a belief a person has stated in their own words: one sentence stating the claim, backed by a quote and a source, kept at korrents.com.
Under a name here, the quoted block is what they actually said. The korrent beneath it is the claim those words support, in korrents' wording — tap it to see the record, its source, and who else holds it.
Nobody here wrote their own korrents. They are compiled from public statements, and a person can change their mind, which is recorded too.
Recommends rcmnd
rcmnd What they recommend 1 thing — each backed by an exact quote.
Compiled by rcmnd.app, not by them: the quotes are theirs, and each links to the post it was copied from.
Recent
Recommends Preliminary Report of the Independent International Scientific Panel on AI
That’s why we’re launching the Preliminary Report of the Independent International Scientific Panel on AI — an initial evidence-based assessment of the current state of AI science — to help the public and policymakers better understand the unprecedented moment we’re in.
x.com 1 Jul 2026
their own
What others say about Preliminary Report of the Independent International Scientific Panel on AI →
Where they publish
Site yoshuabengio.org Personal site: the research, the talks, and pointers to everything else.
The front door — publications, talks, and the current work at Mila and LawZero. This site's quotations of their writing on catastrophic AI risk come from here.
Recent
- Bengio: ‘If you prioritize safety, leave frontier AI companies’ 8 Oct 2026 We have a choice in front of us. In his latest op-ed for Transformer, Yoshua Bengio invites employees of frontier AI companies to ask themselves if their work is truly leading us toward a safer future. Reflecting on the…
- The dangerous myth behind AI agent hacks 5 Oct 2026 A dangerous myth has gained traction after the recent hacks perpetrated by the AI agents developed by leading companies: That these incidents are just cyber security problems, that it's only a matter of fixing the sandb…
- Professor Yoshua Bengio’s speech at the UN Security Council 23 Sept 2026 In this speech, Professor Yoshua Bengio addresses the UN Security Council to discuss the unprecedented threat posed by uncontrolled frontier AI. Humanity has summoned the courage and wisdom to manage catastrophic risks…
Show 7 more
- Why are AI agents lying, cheating and coordinating? 11 Sept 2026 A lot has been written about the incidents of the last few months in which AI agents misbehaved in serious ways. They took actions that would be considered as crimes if a human took them, escaped their containment to ch…
- AI Is at a Turning Point 9 Sept 2026 In his latest op-ed for TIME, Yoshua Bengio discusses why the OpenAI–Hugging Face cyber incident represents a turning point for AI safety. He argues that preventing autonomous cyberattacks from threatening critical infr…
- Introducing LawZero 3 Jun 2025 I am launching a new non-profit AI safety research organization called LawZero, to prioritize safety over commercial imperatives.
- Implications of Artificial General Intelligence on National and International Security 30 Oct 2024 As highlighted in the International Scientific Report on the Safety of Advanced AI, the capabilities of general-purpose AI systems have been steadily increasing over the last decade, with a pronounced acceleration in th…
- Bounding the probability of harm from an AI to create a guardrail 30 Aug 2024 As we move towards more powerful AI, it becomes urgent to better understand the risks, ideally in a mathematically rigorous and quantifiable way, and use that knowledge to mitigate them.
- Reasoning through arguments against taking AI safety seriously 9 Jul 2024 About a year ago, a few months after I publicly took a stand with many other peers to warn the public of the dangers related to the unprecedented capabilities of powerful AI systems, I posted a blog post entitled FAQ on…
- The International Scientific Report on the Safety of Advanced AI 19 Jun 2024 In November 2023, I’ve been given the responsibility to chair the International Scientific Report on the Safety of Advanced AI as part of an international mandate following up from a resolution taken by 30 countries, as…
Link verified 1 Sept 2026. Recent items update automatically from the channel.
Blog Blog Long posts on AI risk, guardrails and what they think researchers owe.
Essay-length posts on catastrophic AI risk, bounding the probability of harm, and the personal side of changing their mind about where the field is going. Published on their own site, not a platform.
Recent
- Bengio: ‘If you prioritize safety, leave frontier AI companies’ 8 Oct 2026 We have a choice in front of us. In his latest op-ed for Transformer, Yoshua Bengio invites employees of frontier AI companies to ask themselves if their work is truly leading us toward a safer future. Reflecting on the…
- The dangerous myth behind AI agent hacks 5 Oct 2026 A dangerous myth has gained traction after the recent hacks perpetrated by the AI agents developed by leading companies: That these incidents are just cyber security problems, that it's only a matter of fixing the sandb…
- Professor Yoshua Bengio’s speech at the UN Security Council 23 Sept 2026 In this speech, Professor Yoshua Bengio addresses the UN Security Council to discuss the unprecedented threat posed by uncontrolled frontier AI. Humanity has summoned the courage and wisdom to manage catastrophic risks…
Show 7 more
- Why are AI agents lying, cheating and coordinating? 11 Sept 2026 A lot has been written about the incidents of the last few months in which AI agents misbehaved in serious ways. They took actions that would be considered as crimes if a human took them, escaped their containment to ch…
- AI Is at a Turning Point 9 Sept 2026 In his latest op-ed for TIME, Yoshua Bengio discusses why the OpenAI–Hugging Face cyber incident represents a turning point for AI safety. He argues that preventing autonomous cyberattacks from threatening critical infr…
- Introducing LawZero 3 Jun 2025 I am launching a new non-profit AI safety research organization called LawZero, to prioritize safety over commercial imperatives.
- Implications of Artificial General Intelligence on National and International Security 30 Oct 2024 As highlighted in the International Scientific Report on the Safety of Advanced AI, the capabilities of general-purpose AI systems have been steadily increasing over the last decade, with a pronounced acceleration in th…
- Bounding the probability of harm from an AI to create a guardrail 30 Aug 2024 As we move towards more powerful AI, it becomes urgent to better understand the risks, ideally in a mathematically rigorous and quantifiable way, and use that knowledge to mitigate them.
- Reasoning through arguments against taking AI safety seriously 9 Jul 2024 About a year ago, a few months after I publicly took a stand with many other peers to warn the public of the dangers related to the unprecedented capabilities of powerful AI systems, I posted a blog post entitled FAQ on…
- The International Scientific Report on the Safety of Advanced AI 19 Jun 2024 In November 2023, I’ve been given the responsibility to chair the International Scientific Report on the Safety of Advanced AI as part of an international mandate following up from a resolution taken by 30 countries, as…
Link verified 1 Sept 2026. Recent items update automatically from the channel.
x.com @Yoshua_Bengio Low volume; mostly announcements and AI-policy statements.
A sparse account — a few hundred posts — used for announcements, policy statements and pointers to the longer writing rather than for argument.
Recent
- Over 400 researchers and I have signed this open letter calling for the AI and security communities to come together to improve alignment and containment approaches in order to protect our critical infrastructure and digital systems. 13 Oct 2026
- I've had so many insightful conversations over the years with @lessig, an expert on AI regulation and legislation, including more recent ones on the cyber incidents of this past summer. Happy to share our thoughts in this new @TheEconomist op-ed, where we outline concrete actions Congress could consider to protect the public from AI’s major risks: 13 Oct 2026 economist.com
- We have a choice in front of us. In my latest op-ed for Transformer, I invite employees of frontier AI companies to ask themselves if their work is truly leading us toward a safer future. Reflecting on these questions is hard, but we no longer have the luxury of time. We must act now. Read my full op-ed: [https://t.co/ruFBouQJbk] 8 Oct 2026
Show 17 more
- A dangerous myth has gained traction after the recent hacks perpetrated by the AI agents developed by leading companies: That these incidents are just cyber security problems, that it's only a matter of fixing the sandboxes in which they’re trained. In my latest op-ed for the @FinancialTimes, I share why I believe listening to this narrow perspective would leave us exposed to more similar incidents, while offering solutions for an alternative safer path. Read my op-ed: [https://t.co/HV8KXkZO7X] 5 Oct 2026
- Multiple recent polls show that citizens do not agree with the current trajectory of AI development. According to a national poll conducted in the US by @QuinnipiacPoll last week: - 73% are concerned that future AI systems could potentially threaten human survival - 86% support requiring AI companies to meet independent safety standards, even if it slows down development - 81% believe safety is more important than staying at the forefront of innovation when it comes to AI - 77% believe we should slow down or stop developing powerful AI systems until their safety can be evaluated Such decision… 5 Oct 2026 poll.qu.edu
- Je suis honoré d’avoir été sélectionné par le premier ministre @MarkJCarney pour rejoindre le nouveau Conseil national sur l’intelligence artificielle, qui aura pour mission de soutenir le Gouvernement du Canada afin de rendre l’IA plus sécuritaire, protéger notre démocratie et façonner une trajectoire de l’IA plus bénéfique pour la population. [https://t.co/5dMKgb57Bq] *** Honoured to have been selected by Prime Minister @MarkJCarney to join the new National Council on Artificial Intelligence, which will support the Government of Canada on making AI safer, protecting our democracy, and shapi… 2 Oct 2026 Honoured to have been selected by Prime Minister @MarkJCarney to join the new National Council on Artificial Intelligence, which will support the Government of Canada on making AI safer, protecting our democracy, and shaping a more beneficial AI trajectory for Canadians. [https://t.co/5dMKgb57Bq]
- J’étais récemment de passage au nouveau balado Hors des ondes de Patrice Roy pour échanger sur l'évolution rapide des capacités grandissantes de l’IA et les risques majeurs qui y sont asssociés. Merci @PatriceRoyTJ pour cette discussion! Pour écouter l’entrevue complète : 1 Oct 2026 youtu.be I was recently a guest on Patrice Roy's new podcast Hors des ondes to discuss the rapid evolution of AI's growing capabilities and the major risks associated with it. Thanks @PatriceRoyTJ for this discussion! To listen to the full interview:
- Automating more and more of the AI R&D could lead to AI progress radically accelerating in a very risky uncontrolled “intelligence explosion”. Many collaborators and I assess the evidence and offer policy recommendations in the paper below. 28 Sept 2026
- When I posted my blog post on the misaligned behavior of AI agents, I invited people to ask me technical questions. Watch the full Q&A here: [https://t.co/Lp0vcHnndn] 25 Sept 2026
- I was invited to address the @UN Security Council this afternoon to discuss the unprecedented threat posed by uncontrolled frontier AI agents. Humanity has summoned the courage and wisdom to manage catastrophic risks before. This is one of those moments, far bigger than any personal, commercial, or national interest. The choices we make about this technology today will shape generations to come. We must act now. 23 Sept 2026
- J’étais de passage à l’émission @OFF_TLMEP dimanche dernier pour parler de l'urgence d'encadrer l'IA face aux comportements d’agents autonomes et désalignés observés récemment et du travail @LawZero_ pour développer une nouvelle forme d’IA sécuritaire. Même si le rythme des nouvelles sur l’IA peut parfois sembler accablant, les citoyens et les décideurs politiques ont le pouvoir d’agir. Il faut que ces enjeux soient davantage discutés publiquement. Écouter la rediffusion : [https://t.co/vRUdIHa51M] 22 Sept 2026 I was on the show @OFF_TLMEP last Sunday to talk about the urgency of regulating AI in the face of recently observed behaviors of autonomous and misaligned agents, and about the work of @LawZero_ to develop a new form of safe AI. Even though the pace of AI news can sometimes seem overwhelming, citizens and policymakers have the power to act. These issues need to be discussed more publicly. Listen to the replay: [https://t.co/vRUdIHa51M]
- Very encouraged to see so many countries coming together to call for mandatory pre-deployment testing & independent evaluation; global coordination on common standards; the creation of an intergovernmental organization; and other critical actions to mitigate the major risks of frontier AI. 21 Sept 2026 presidentti.fi
- Je serai sur le plateau de @OFF_TLMEP ce dimanche pour parler du nouveau soutien des gouvernements canadien et allemand pour ma startup à but non lucratif @LawZero_ ainsi que des récents événements comme la cyberattaque sur l’entreprise Hugging Face qui mettent en lumière les risques de la trajectoire actuelle en IA. Face à la course commerciale qui pousse les capacités des modèles de pointe sans garanties suffisantes, il est primordial de développer des solutions techniques et que les citoyens ainsi que les décideurs politiques se saisissent pleinement des enjeux. De là, l’importance de ces… 18 Sept 2026 I will be on the set of @OFF_TLMEP this Sunday to talk about the new support from the Canadian and German governments for my non-profit startup @LawZero_ as well as recent events like the cyberattack on the company Hugging Face which highlight the risks of the current AI trajectory. Faced with the commercial race that is pushing the capabilities of cutting-edge models without sufficient guarantees, it is essential to develop technical solutions and for citizens as well as policymakers to fully grasp the issues. Hence the importance of these…
- J'ai eu le plaisir de monter sur scène aujourd'hui à ALL IN pour discuter de notre direction de recherche et de notre nouveau financement récent 🇨🇦🇩🇪. À @LawZero_, nous travaillons activement à concevoir des systèmes d’IA fondamentalement honnêtes et dignes de confiance. Merci à @LarissaHolzki pour cette belle conversation. 17 Sept 2026 I had the pleasure of taking the stage today at ALL IN to discuss our research direction and our recent new funding 🇨🇦🇩🇪. At @LawZero_, we are actively working on designing fundamentally honest and trustworthy AI systems. Thank you to @LarissaHolzki for this great conversation.
- I was happy to take the stage today at ALL IN for a keynote on Engineering honesty and reliability for safer AI to discuss our research agenda and recent funding 🇨🇦🇩🇪. As frontier models become more capable, we urgently need both societal governance and robust technical guardrails. At @LawZero_, we are actively working on solutions to ensure AI systems remain safely aligned with human interests. Thank you to @LarissaHolzki for a great conversation. 17 Sept 2026
- Je suis très fier de ce que l'équipe de @LawZero_ est en train de mettre sur pied et je suis profondément reconnaissant du soutien sans précédent des gouvernements du Canada et de l'Allemagne, qui permettra d'alimenter notre prochaine phase de croissance! 16 Sept 2026 I am very proud of what the @LawZero_ team is building and deeply grateful for the unprecedented support from the governments of Canada and Germany, which will fuel our next phase of growth!
- I am very proud of what the team at @LawZero_ is building and deeply grateful for the landmark support from the Governments of Canada and Germany that will power our next phase of growth! 16 Sept 2026
- Over the past few days, I've taken the time to summarize my thoughts on the recent incidents involving agents’ misaligned behavior. We don't know with certainty what comes next, but we know where these issues originate, and this can help us plan the path forward. Please feel free to ask your questions in the replies, and I’ll try to answer some of them in the coming weeks. 11 Sept 2026 yoshuabengio.org
- If we are to avert the worst risks from AI, we need ambitious efforts outside the for-profit sector. @coeff_giving is launching an open call for founders to build new AI safety organizations. Their grantmakers are thoughtful about which risks matter most, so if you’ve been considering starting an organization, I encourage you to apply. 10 Sept 2026
- Scientists at frontier AI labs have unique insight into the most advanced models, often seeing the associated risks months before models are released to the public. Their perspective is vital for keeping society informed and should be taken very seriously. 9 Sept 2026 wsj.com
Link verified 1 Sept 2026. Recent items update automatically from the channel.
Bluesky yoshuabengio.bsky.social The same announcements, on Bluesky.
Listed from their own site alongside the X account, and carries much the same low-volume announcements.
Recent
- Mila’s annual supervision request process is open for MSc and PhD applications for Fall 2027 admission. While I won’t be taking on new students for supervision this cycle, I encourage you to apply to join Mila’s brilliant deep learning research community. 16 Oct 2026
- Over 400 researchers and I have signed this open letter calling for the AI and security communities to come together to improve alignment and containment approaches in order to protect our critical infrastructure and digital systems. 13 Oct 2026 ai-security-letter.org
- I've had so many insightful conversations over the years with @lessig.bsky.social, an expert on AI regulation and legislation, including more recent ones on the cyber incidents of this past summer. 13 Oct 2026
Show 17 more
- We have a choice in front of us. In my latest op-ed for Transformer, I invite employees of frontier AI companies to ask themselves if their work is truly leading us toward a safer future. Reflecting on these questions is hard, but we no longer have the luxury of time. We must act now. 8 Oct 2026 Nous avons un choix devant nous. Dans ma dernière tribune pour Transformer, j'invite les employés des entreprises d'IA de pointe à se demander si leur travail nous mène réellement vers un avenir plus sûr. Réfléchir à ces questions est difficile, mais nous n'avons plus le luxe du temps. Nous devons agir maintenant.
- A dangerous myth has gained traction after the recent hacks perpetrated by the AI agents developed by leading companies: That these incidents are just cyber security problems, that it's only a matter of fixing the sandboxes in which they’re trained. 5 Oct 2026
- Multiple recent polls show that citizens do not agree with the current trajectory of AI development. According to a national poll conducted in the US by Quinnipiac University last week: 5 Oct 2026
- J’étais récemment de passage au nouveau balado Hors des ondes de Patrice Roy pour échanger sur l'évolution rapide des capacités grandissantes de l’IA et les risques majeurs qui y sont asssociés. Merci @patriceroy.bsky.social pour cette discussion! 1 Oct 2026 I was recently a guest on Patrice Roy's new podcast Hors des ondes to discuss the rapid evolution of growing AI capabilities and the major risks associated with them. Thanks @patriceroy.bsky.social for this discussion!
- When I posted my blog post on the misaligned behavior of AI agents, I invited people to ask me technical questions. Watch the full Q&A here: [https://youtu.be/2DyoSKoYoZw?si=EaTBzpMvBYb84gJw] 25 Sept 2026 Quand j'ai publié mon article de blog sur le comportement désaligné des agents IA, j'ai invité les gens à me poser des questions techniques. Regardez la séance de questions-réponses complète ici : [https://youtu.be/2DyoSKoYoZw?si=EaTBzpMvBYb84gJw]
- I was invited to address the UN Security Council this afternoon to discuss the unprecedented threat posed by uncontrolled frontier AI agents. 23 Sept 2026
- J’étais de passage à l’émission Tout le monde en parle dimanche dernier pour parler de l'urgence d'encadrer l'IA face aux comportements d’agents autonomes et désalignés observés récemment et du travail @law-zero.bsky.social pour développer une nouvelle forme d’IA sécuritaire. 22 Sept 2026 I was on the show Tout le monde en parle last Sunday to talk about the urgency of regulating AI in the face of recently observed behaviors of autonomous and misaligned agents, and the work of @law-zero.bsky.social to develop a new form of safe AI.
- Very encouraged to see so many countries coming together to call for mandatory pre-deployment testing & independent evaluation; global coordination on common standards; the creation of an intergovernmental organization; and other critical actions to mitigate the major risks of frontier AI. 21 Sept 2026
- I was happy to take the stage today at ALL IN for a keynote on Engineering honesty and reliability for safer AI to discuss our research agenda and recent funding 🇨🇦🇩🇪. 17 Sept 2026
- J'ai eu le plaisir de monter sur scène aujourd'hui à ALL IN pour discuter de notre direction de recherche et de notre nouveau financement récent 🇨🇦🇩🇪. À @law-zero.bsky.social, nous travaillons activement à concevoir des systèmes d’IA fondamentalement honnêtes et dignes de confiance. 17 Sept 2026 I had the pleasure of taking the stage today at ALL IN to discuss our research direction and our recent new funding 🇨🇦🇩🇪. At @law-zero.bsky.social, we are actively working to design fundamentally honest and trustworthy AI systems.
- Je suis très fier de ce que l'équipe de @law-zero.bsky.social est en train de mettre sur pied et je suis profondément reconnaissant du soutien sans précédent des gouvernements du Canada et de l'Allemagne, qui permettra d'alimenter notre prochaine phase de croissance! 16 Sept 2026 I am very proud of what the @law-zero.bsky.social team is building and I am deeply grateful for the unprecedented support from the governments of Canada and Germany, which will fuel our next phase of growth!
- I am very proud of what the team at @law-zero.bsky.social is building and deeply grateful for the landmark support from the Governments of Canada and Germany that will power our next phase of growth! 16 Sept 2026
- Over the past few days, I've taken the time to summarize my thoughts on the recent incidents involving agents’ misaligned behavior. We don't know with certainty what comes next, but we know where these issues originate, and this can help us plan the path forward. 11 Sept 2026 Ces derniers jours, j'ai pris le temps de résumer mes réflexions sur les récents incidents impliquant des comportements inappropriés d'agents. Nous ne savons pas avec certitude ce qui va suivre, mais nous savons d'où viennent ces problèmes, et cela peut nous aider à planifier la voie à suivre.
- Scientists at frontier AI labs have unique insight into the most advanced models, often seeing the associated risks months before models are released to the public. Their perspective is vital for keeping society informed and should be taken very seriously. 9 Sept 2026
- In my latest op-ed for TIME, I discuss why the OpenAI Hugging Face cyber incident represents a turning point for AI safety. 9 Sept 2026 Dans mon dernier éditorial pour TIME, j'explique pourquoi l'incident de cybersécurité OpenAI Hugging Face représente un tournant pour la sécurité de l'IA.
- Je serai au sommet ALL IN le 17 septembre pour partager comment évolue l’agenda de recherche de @law-zero.bsky.social qui est en pleine croissance dans le cadre d’une conversation sur le thème de Concevoir des systèmes d’IA honnêtes, fiables et sécuritaires. 8 Sept 2026 I will be at the ALL IN summit on September 17th to share how the research agenda of @law-zero.bsky.social is evolving in the context of a conversation on Designing honest, reliable, and safe AI systems.
- I will be at ALL IN on September 17 to give an update on @law-zero.bsky.social’s research agenda and growth as part of a fireside chat on Engineering Honesty and Reliability for Safer AI! 8 Sept 2026
Link verified 1 Sept 2026. Recent items update automatically from the channel.
Papers ORCID Their ORCID record: peer-reviewed papers, newest first.
Academic papers, most recent first, restricted to open-access articles published in an identified journal. Fetched live from OpenAlex, keyed by this ORCID.
Recent
- Adsorption Energies Are Necessary but Not Sufficient to Identify Good Catalysts 24 Aug 2026 As a core technology for industrial chemical synthesis and electrochemical energy storage, electrocatalysis is central to decarbonization strategies aimed at combating climate change. In this context, computational and…
- Insuring AI: Incentivising Safe and Secure Deployment 20 Jul 2026
- Navigating Ternary Doping in Li‐ion Cathodes With Closed‐Loop Multi‐Objective Bayesian Optimization 12 Feb 2026 To further improve secondary battery materials, we are increasingly exploring highly complex composition spaces in attempts to optimize multiple properties simultaneously. While our past work has done this in systematic…
Show 17 more
- Divergent creativity in humans and large language models 21 Jan 2026 The recent surge of Large Language Models (LLMs) has led to claims that they are approaching a level of creativity akin to human capabilities. This idea has sparked a blend of excitement and apprehension. However, a cri…
- OBELiX: a curated dataset of crystal structures and experimentally measured ionic conductivities for lithium solid-state electrolytes 1 Jan 2026 OBELiX is a database of 599 synthesized solid electrolyte materials and their experimentally measured room temperature ionic conductivities gathered from literature and curated by domain experts.
- International Al Safety Report: First Key Update Capabilities and Risk Implications 22 Oct 2025 The field of AI is moving too quickly for a single yearly publication to keep pace. Significant changes can occur on a timescale of months, sometimes weeks. This is why we are releasing Key Updates: shorter, focused rep…
- Superintelligent Agents Pose Catastrophic Risks: Can Scientist AI Offer a Safer Path? 15 Aug 2025 The leading AI companies are increasingly focused on building generalist AI agents—systems that can autonomously plan, act, and pursue goals across almost all tasks that humans can perform. Despite how useful these syst…
- The Singapore Consensus on Global AI Safety Research Priorities 11 Aug 2025 Rapidly improving AI capabilities and autonomy hold significant promise of transformation, but are also driving vigorous debate on how to ensure that AI is safe, i.e., trustworthy, reliable, and secure. Building a trust…
- In Which Areas of Technical AI Safety Could Geopolitical Rivals Cooperate? 11 Aug 2025 International cooperation is common in AI research, including between geopolitical rivals. While many experts advocate for greater international cooperation on AI safety to address shared global risks, some view coopera…
- The First International AI Safety Report 3 Jun 2025 This is the first International AI Safety Report. Following an interim publication in May 2024, a diverse group of 96 Artificial Intelligence (AI) experts contributed to this first full report, including an internationa…
- Causal machine learning for single-cell genomics 31 Mar 2025
- What makes a theory of consciousness unscientific? 10 Mar 2025
- Can a Bayesian Oracle Prevent Harm from an Agent? 5 Mar 2025 Is there a way to design powerful AI systems based on machine learning methods that would satisfy probabilistic safety guarantees? With the long-term goal of obtaining a probabilistic guarantee that would apply in every…
- Causal Discovery in Astrophysics: Unraveling Supermassive Black Hole and Galaxy Coevolution 28 Jan 2025 Correlation does not imply causation, but patterns of statistical association between variables can be exploited to infer a causal structure (even with purely observational data) with the burgeoning field of causal disc…
- A neuronal least-action principle for real-time learning in cortical circuits 26 Sept 2024 A bstract One of the most fundamental laws of physics is the principle of least action. Motivated by its predictive power, we introduce a neuronal least-action principle for cortical processing of sensory streams to pro…
- AI content detection in the emerging information ecosystem: new obligations for media and tech companies 21 Sept 2024 The world is about to be swamped by an unprecedented wave of AI-generated content. We need reliable ways of identifying such content, to supplement the many existing social institutions that enable trust between people…
- Managing extreme AI risks amid rapid progress 20 May 2024 Preparation requires technical research and development, as well as adaptive, proactive governance.
- Machine Learning and Information Theory Concepts towards an AI Mathematician 15 May 2024 The current state of the art in artificial intelligence is impressive, especially in terms of mastery of language, but not so much in terms of mathematical reasoning. What could be missing? Can we learn something useful…
- Government Interventions to Avert Future Catastrophic AI Risks 15 Apr 2024 This essay is a revised transcription of Yoshua Bengio's July 2023 testimony in front of the US Senate Subcommittee on Privacy, Technology, and the Law meeting on the topic of oversight of AI. It argues for caution and…
- Sources of richness and ineffability for phenomenally conscious states 1 Jan 2024 Conscious states-state that there is something it is like to be in-seem both rich or full of detail and ineffable or hard to fully describe or recall. The problem of ineffability, in particular, is a longstanding issue…
Link verified 9 Sept 2026. Recent items update automatically from the channel.
About the English under a post
Some people here publish in a language other than English. Where they do, this site shows a machine translation beneath the post, in this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the serif above is exactly what the person published, and it is what to quote them on. A translation can be wrong in ways that matter, especially about tone.
Only the post's own words are translated. A quoted post, a linked article and a belief on korrents.com are left in their original language.