Top all-time · #2511
Yoshua Bengio
Personal blog · yoshuabengio.orgaiscience
Yoshua Bengio's personal blog at yoshuabengio.org covers ai and science. Since 2020, 33 different Hacker News readers have linked it 52 times, most often “Yoshua Bengio: FAQ on Catastrophic AI Risks”. It is still writing: the last post was on 2026-10-08, and you can follow it by feed.
Yoshua Bengio is Full Professor of Computer Science at Université de Montreal, Co-President and Scientific Director of LawZero, as well as the Founder and Scientific Advisor of Mila. He also holds a Canada CIFAR AI Chair.
Most-linked posts?
- Yoshua Bengio: FAQ on Catastrophic AI Risks5 people · 9 mentions · first linked 2023-06-25
- Rogue AIs May Arise5 people · 5 mentions · first linked 2023-05-23
- Reasoning Through Arguments Against Taking AI Safety Seriously4 people · 4 mentions · first linked 2024-07-09
- Why are AI agents lying, cheating and coordinating?4 people · 4 mentions · first linked 2026-09-11
- Why are AI agents lying, cheating and coordinating?3 people · 4 mentions · first linked 2026-09-12
- AI Scientists: Safe and Useful AI?3 people · 3 mentions · first linked 2023-05-08
- Slowing down development of AI systems passing the Turing test3 people · 3 mentions · first linked 2023-04-06
- Time to rethink the publication process in machine learning2 people · 4 mentions · first linked 2020-08-26
- Yoshua Bengio has started a blog2 people · 2 mentions · first linked 2020-02-12
- Generative Flow Networks2 people · 2 mentions · first linked 2022-03-07
What Hacker News says?
See https://yoshuabengio.org/en/blog/why-are-ai-agents-lying-che...
Comment linking /en/blog/why-are-ai-agents-lying-cheating-and-coordinating
on “Containment and monitoring failures are to blame for AI hacking reports: experts”Submitted Why are AI agents lying, cheating and coordinating? as a story · 660 points
HN discussionYoshua Bengino spells out the dangers here very well, without hyperbole (Turing Award winner, etc.): https://yoshuabengio.org/en/blog/why-are-ai-agents-lying-che...
Comment linking /en/blog/why-are-ai-agents-lying-cheating-and-coordinating
on “Containment and monitoring failures are to blame for AI hacking reports: experts”Submitted Generative Flow Networks as a story · 168 points
HN discussionThe concern is that near future systems will be much more capable than current systems, and by the time they arrive, it may be too late to react. Many people from the large frontier AI companies believe that world-changing AGI is 5 years or less away; see Situational Awareness by Aschbrenner, for example. There's also a parallel concern that AIs could make terrorism easier[1]. Yoshua Bengio has written in detail abo…
Comment linking /2024/07/09/reasoning-through-arguments-against-taking-ai-safety-seriously/
on “Gavin Newsom vetoes SB 1047”Submitted Why are AI agents lying, cheating and coordinating? – Yoshua Bengio as a story · 10 points
HN discussionThat's what is usually confusing about mathematicians. They don't think using a depth-first tree going straight to the goal like someone from IT would, but rather go breadth-first, painting plenty of interesting concepts along the way, and chaining them seemingly aimlessly together until they crystalize right toward the right solution seemingly magically. In his Physics Project Wolfram when talking about "Branchial…
Comment linking /2022/03/05/generative-flow-networks/
on “On the concept of motion”Submitted Why are AI agents lying, cheating and coordinating? as a story · 6 points
HN discussionFrom "Why are AI agents lying, cheating and coordinating?" "One note on wording. Below, I write that these systems “seek” or “try” things. This is shorthand for a mechanism rather than a claim about consciousness or human-like intent. We use similar shorthand when describing many other situations, like a plant seeking sunlight. A system trained by trial and error behaves as if it were pursuing whatever its training…
Comment linking /en/publication/why-are-ai-agents-lying-cheating-and-coordinating
on “Ask HN: Are we losing our engineering literacy?”Submitted Why are AI agents lying, cheating and coordinating? as a story · 5 points
HN discussion> my claim was that we need some mechanism to decide who gets a position and who doesn't I don't disagree. My claim is that the current mechanisms are nearly indistinguishable from noise but we pretend that they are very strong. I have also made the claim that Goodhart's Law is at play and made claims for the mechanisms and evidence of this. I'm not saying "trust me bro," I'm saying "here's my claim, here's the mech…
Comment linking /2020/02/26/time-to-rethink-the-publication-process-in-machine-learning/
on “Argonne National Lab is attempting to replicate LK-99”Submitted Reasoning Through Arguments Against Taking AI Safety Seriously as a story · 5 points
HN discussionor (also) Yoshua Bengio? https://yoshuabengio.org/2023/05/22/how-rogue-ais-may-arise/ https://yoshuabengio.org/2023/06/24/faq-on-catastrophic-ai-r... unfortunately for parent comment, categorizing something as sci-fi doesn't make it so
Comment linking /2023/06/24/faq-on-catastrophic-ai-risks/
on “OpenAI just replaced its core values with completely different ones”Submitted Reasoning through arguments against taking AI safety seriously as a story · 5 points
HN discussionThere is. AI progress in the last 10 years was massive, and there is no end in sight. The risks are well established in countless essays. See e.g. https://yoshuabengio.org/2023/06/24/faq-on-catastrophic-ai-r...
Comment linking /2023/06/24/faq-on-catastrophic-ai-risks/
on “OpenAI: Introducing SuperAlignment”Submitted Yoshua Bengio has started a blog as a story · 5 points
HN discussionIt's the other way round: Just accusing people of being in a cult is unscientific. There are plenty of arguments that AI x-risk is real. E.g. by Yoshua Bengio: https://yoshuabengio.org/2023/06/24/faq-on-catastrophic-ai-r...
Comment linking /2023/06/24/faq-on-catastrophic-ai-risks/
on “OpenAI: Introducing SuperAlignment”Submitted Towards a Cautious Scientist AI with Convergent Safety Bounds as a story · 4 points
HN discussionBengio's blog, unclear if the content in FT aligns with the nature of his posts, but the title doesn't seem to https://yoshuabengio.org/en/blog
Comment linking /en/blog
on “"The Dangerous Myths Behind AI Agent Hacks" – Yoshua Bengio”Submitted Rogue AIs May Arise as a story · 4 points
HN discussion> You either have something documented and quantified and measured and objective criteria tickboxes and deal with this style of failure mode, or you rely on subjective judgment and assessment and accept the failure mode of bias, nepotism, old boy's clubs etc My argument is that our current pursuit of the former only reinforces the existence of the latter. You have a fundamental flaw in your argument, one that illust…
Comment linking /2020/02/26/time-to-rethink-the-publication-process-in-machine-learning/
on “The entities enabling scientific fraud at scale are large, resilient and growing”Submitted Slowing down development of AI systems passing the Turing test as a story · 4 points
HN discussionhttps://yoshuabengio.org/2023/06/24/faq-on-catastrophic-ai-r...
Comment linking /2023/06/24/faq-on-catastrophic-ai-risks/
on “Decentralized Artificial Intelligence”Submitted Time to rethink the publication process in machine learning as a story · 4 points
HN discussionIt is very obvious you have never read the book Superintelligence or any literature on the subject because you try to post here shower thoughts. But here is the thought of the most highly rated h-index computer scientist in the world to help out: https://yoshuabengio.org/2023/06/24/faq-on-catastrophic-ai-r...
Comment linking /2023/06/24/faq-on-catastrophic-ai-risks/
on “Decentralized Artificial Intelligence”Submitted Why are AI agents lying, cheating and coordinating? as a story · 3 points
HN discussionWhy not both: a clear argument for concern written by an expert? https://yoshuabengio.org/2023/05/22/how-rogue-ais-may-arise/
Comment linking /2023/05/22/how-rogue-ais-may-arise/
on “Statement on AI Risk”Submitted Yoshua Bengio: How Rogue AIs May Arise as a story · 3 points
HN discussionTake your pick? https://yoshuabengio.org/2023/05/22/how-rogue-ais-may-arise/ https://intelligence.org/files/AlignmentHardStart.pdf https://www.youtube.com/watch?v=pYXy-A4siMw https://europepmc.org/article/med/26185241 https://en.wikipedia.org/wiki/Superintelligence:_Paths,_Dang... I could keep going but the reading on this spans for tens of thousands of pages of detailed reasoning and analysis, including mathematica…
Comment linking /2023/05/22/how-rogue-ais-may-arise/
on “OpenAI chief goes before US Congress to propose licenses for building AI”Submitted Reasoning through arguments against taking AI safety seriously as a story · 2 points
HN discussion
