Knowledge catalogue
woah @Noahpinion is attacking @demishassabis here — calling what Demis said not “credible” — but misinterpreting what Hassabis is *actually* saying. What Demis actually said is both completely credibl
AGI is certainly not here by the definitions I have repeatedly laid out. or by criteria that @hendrycks @Yoshua_Bengio and I and others recently laid about at http://agidefinition.AI i don’t think any
Gary Marcus posted a clarification about humor and irony in a previous post that was apparently misunderstood by readers, indicating he felt the need to make his intended meaning more explicit. The po
Can't agree more. After trying out ChatGPT 5.4/5.5 for two months I feel it's far from replacing human researchers, even though it's enabling me to do things not possible before. The effect on how out
fascinating - the X algorithm literally gives no weight to being truthful or well-sourced. which kind of shows. So I spent some time studying the new Twitter/X algorithm today since the latest version
Gary has been wrong about plenty of things, and right about plenty more. That's what happens when you say things publicly about the future. I disagree with him about lots of his expectations about the
This appears to be a social media comment acknowledging Gary Marcus's previous critical positions on AI, with the commenter suggesting that negative reactions directed at Marcus validate the accuracy
This post from Gary Marcus appears to reference a humorous or notable moment related to his work, with image attachments in Traditional Chinese that provide context to the claim that he 'has won again
@GaryMarcus @hendrycks @Yoshua_Bengio We recently tested all of the major LLMs with tic-tac-toe, modified chess, and a novel game -- even the top models all failed: illegal moves, claiming a win when
GOOGLE CEO: “THERE IS SOME IRRATIONALITY IN THE CURRENT AI BOOM. THE GROWTH OF AI INVESTMENT HAS BEEN AN EXTRAORDINARY MOMENT.” ASKED WHAT HAPPENS IF THE AI BUBBLE POPS, HE SAID: “I THINK NO COMPANY I
have always wondered about this: is it that the true believers willfully simply don’t want to understand my actual positions or do they lack the cognitive chops to follow them? @GaryMarcus This, sadly
here’s one sample, below, not even the only one I saw today; another claimed I don’t do research anymore when I had published articles in two top journals (Science and the Proceedings of The Royal Soc
he’s not dumb and certainly his IQ is above average but as I wrote somewhere before “I think that Musk has consistently made a dumb set of choices - largely stemming from a mix of four cognitive error
hilarious how all the objections to this post about how elon understands engineering better than science point to elon’s engineering accomplishments, rather than his (meager) record of scientific inno
⚠️⚠️⚠️i don’t think most people understand the implications of the mood shift below, so I will spell them out. they are serious, and eventually will affect the global economy. when a serious coder as
> I’m now in the LeCun/Marcus camp on LLMs > real programming agents will need world models > not some RLVR shit it’s over The Eternal Sloptember https://geohot.github.io//blog/jekyll/update/2026/05/2
Interesting point … but are there any critics of mine that are “authentic” who have laid out a systematic critique, without relying on strawmen? @GaryMarcus While addressing some critics can be helpfu
interestingly (although it is bit apples and oranges) the number below is a similar to the fraction of cars involved in car accidents (mostly nonfatal) in a week. but, crucially, we have tons of measu
Let the record show: honest critique is not pessimism. Your points on scaling limits and neurosymbolic AI have been validated. Fair debate requires engaging with what was actually said. Retweeting thi
LLM companies are likely to be like airline companies: small margins, intense competition, high expenses. @GaryMarcus Airlines taught us something important: unlimited demand does not guarantee attrac
Gary Marcus discusses whether it is worthwhile to publicly document and explain the logical and conceptual errors made by his critics, framing it as a potential educational exercise. The post reflects
neurosymbolic by @swarat et al for the Erdos win, with much more careful, quantitative work than openai’s in hindsight i wonder whether OpenAI rushed theirs out, knowing this was coming? Another 9 ope
not using LLM’s works for me Prolonged AI use may make it harder to think critically and creatively, recent research suggests. But there are ways to keep the brain fit https://www.economist.com/scienc
Gary Marcus critiques Elon Musk's public perception, arguing that those lacking scientific expertise tend to overestimate Musk's genius while scientifically literate individuals recognize significant
sure, but could you look up at the sky with utter sincerity while you said it? @GaryMarcus 'Fastest growing business', 'loses $5B a year'. I could easily hemorrhage at least 3x that, if that's what 'b
The desperation of the tech bros is made vivid in all the lies they tell about me. Lies about my training. Lies about my experience. Lies about my publication record. Lies about the predictions I have
A Wall Street Journal report highlights the paradox that while the artificial intelligence industry experiences rapid growth, American public sentiment toward AI is increasingly negative. The article,
these kind of examples were cute three years ago now they are just sad 🦔Four viral examples of Google's AI Overview misfiring hit social media this week. One user searching for the definition of 'disr
“We will have a crash, I just can't tell you when, and I can't tell you how deep. But I can assure you, unfortunately, I wish I wasn't saying this, we will have a crash,” says Andrew Ross Sorkin, fina
Well over 300,000 people seem to have felt the same way, roon’s attack, arrogant, misleading, and unprovoked, did OpenAI no favors. @GaryMarcus roon's attack on you is really making me think less of h
100%: if OpenAI flops, the gravy train stops. @GaryMarcus if OpenAI's IPO flops, gravy train is OVER! whole industry knows it, they're terrified of it, particularly as they're starting to get just how
arXiv:2605.22622v1 Announce Type: new Abstract: Wasserstein Policy Optimization (WPO) is a recently proposed reinforcement learning algorithm that leverages Wasserstein gradient flows to optimize stoc
arXiv:2411.08126v2 Announce Type: replace-cross Abstract: We study offline dynamic pricing when historical data provide incomplete coverage of the price space such that some candidate prices, includin
Gary Marcus expressed concern about what he perceived as extreme desperation among those involved in IPO (Initial Public Offering) markets, characterizing their behavior as the most desperate he had w
arXiv:2605.22155v1 Announce Type: new Abstract: Symbolic methods are generally not considered competitive with strong modern learners on realistic supervised tasks. We evaluate Algebraic Machine Learn
another of the dudes going after me (Theo) is perhaps paid by OpenAI 🙄 He doesn’t work for them, but he likely has or has had some paid partnership with them, because they’ve invited him onto their ca
arXiv:2605.22222v1 Announce Type: new Abstract: Partial differential equation (PDE) foundation models are pretrained networks that forecast how physical fields like velocity and pressure evolve from a
arXiv:2603.02938v2 Announce Type: replace Abstract: Graph-based tasks in the zero-shot setting remain a significant challenge due to data scarcity and the inability of traditional Graph Neural Network
arXiv:2605.22468v1 Announce Type: new Abstract: Cross-subject generalization in biomedical time-series refers to training on data from some subjects and testing on unseen subjects.The key challenge is
arXiv:2605.21846v1 Announce Type: cross Abstract: Causal discovery from multivariate time series is challenging when causal effects may occur both across time and within the same sampling interval. Th
arXiv:2605.21915v1 Announce Type: cross Abstract: Congestion controllers (CCs) are critical to network performance, and yet their robustness under adverse conditions remains insufficiently understood.
arXiv:2509.08933v2 Announce Type: replace Abstract: We study the problem of learning the optimal policy in a discounted, infinite-horizon reinforcement learning (RL) setting in the presence of adversa
arXiv:2605.22401v1 Announce Type: new Abstract: Does the relationship between learning rules and brain alignment generalize across species? We extend our prior finding that untrained CNNs match backpr
arXiv:2510.00319v2 Announce Type: replace Abstract: Large Language Models (LLMs) have been demonstrating strong reasoning capability with their chain-of-thoughts (CoT), which are routinely used by hum
arXiv:2605.22454v1 Announce Type: new Abstract: Data rehearsal has emerged as a leading approach for mitigating catastrophic forgetting in Continual Reinforcement Learning (CRL). However, existing wor
During his second term, Trump will have cut 2 of the US most powerful innovation and wealth creation engines: 1. skilled legal immigration 2. (non-defense) research budgets We won't see the effect of
arXiv:2605.21993v1 Announce Type: cross Abstract: Ranking systems used in decision-support settings should not only order candidates but also expose evidence that can be independently checked. We stud
arXiv:2605.21563v1 Announce Type: new Abstract: Recent reviews find that the vast majority of published healthcare federated learning (FL) studies never reach real-world deployment. We developed an em
arXiv:2605.21617v1 Announce Type: new Abstract: Inference from interaction maps, such as centromere identification from genome-wide chromosome conformation capture techniques -- notably Hi-C -- can be
arXiv:2605.22596v1 Announce Type: new Abstract: Robotic tasks are typically specified by a tuple of factors, such as the object to be grasped, the obstacles to be avoided, the color of the target, and
arXiv:2605.22340v1 Announce Type: new Abstract: Single-cell RNA sequencing (scRNA-seq) provides high-dimensional profiles of cellular states, enabling data-driven modeling of cellular dynamics over ti
@GaryMarcus If OpenAI gets bailed out, I will lobby with all of my effort to ensure whatever political faction was responsible gets voted out. I'm about to become a single issue voter over AI because
arXiv:2605.22507v1 Announce Type: new Abstract: We propose a new framework for generative modeling based on a discrete-time stochastic control formulation of measure transport. Adapting classic result
arXiv:2605.21516v1 Announce Type: new Abstract: Harness engineering has emerged as an important inference-time technique for large language model (LLM) agents, aiming to improve long-term performance
arXiv:2602.05286v3 Announce Type: replace Abstract: Healthcare facility visit prediction is essential for optimizing healthcare resource allocation and informing public health policy. Despite advanced
Here’s a key line in this mythos update. This is precisely an example of why engineers don’t go away, ever. We’ve made it far easier to create and find security issues, which means the new bottleneck
arXiv:2603.02604v2 Announce Type: replace Abstract: We introduce Heterogeneous Agent Collaborative Reinforcement Learning (HACRL), a new Reinforcement Learning from Verifiable Reward (RLVR) problem th
I am deleting a question I had about @theo in light of his clarification. I hope that he and others will reconsider his/their own slander in light of my own clarifications, such as the one below that
I love that this post, my most foulmouthed ever, has gotten over 100k views. You want to know why it has spread? Simple. Because a lot of people have become really fed up with OpenAI. Fuck this OpenAI
Gary Marcus expresses concern that AI-generated low-quality mathematical content ('AI slop') will proliferate, making the role of professional mathematicians increasingly important for quality control