ClaimsPeople

Gary Marcus

Gary Marcus is an author and AI entrepreneur and professor emeritus at NYU; he has written six books, including Rebooting AI and Taming Silicon Valley, founded Geometric Intelligence (bought by Uber), testified at the May 2023 US Senate AI hearing, and writes a newsletter with over 121,000 subscribers.

712 claims collected, 213 researched, 103 resolved. Report of 4 November 2026, method 0.1.0+c876030c.

Scores by kind of claim

Each kind of claim is scored on its own, out of 100: the share that held up, with the range it likely falls in. Fewer resolved claims give a wider range.

  • Factual claims

    Statements about the present or the past, checked against the record: did they hold up?

    79 72 85

    High confidence, 64 factual claims resolved. From 18 documents, 2019 to 2026; 19 documents sampled from 37 candidates found.

    Not scored: 32 unresolvable, 4 checkable only against their own records.

  • Predictions

    Statements about events outside the person's control, checked once they fell due: did they come true?

    80 72 88

    Medium confidence, 38 predictions resolved. From 7 documents, 2019 to 2025; 19 documents sampled from 37 candidates found.

    Not scored: 60 unresolvable. 13 more fall due later.

    Does not yet allow for how likely each prediction was when made.

  • Promises

    Commitments to the person's or organisation's own future action, with something to deliver: was it delivered?

    Insufficient record: 1 promise resolved, 5 needed for a score.

    Not scored: 1 unresolvable.

All types combined: 79 (likely 74 to 84, high confidence), from 102 claims: 64 factual claims, 38 predictions. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.

Gary Marcus's claims

712 claims, sorted by newest said. Choose a claim to read it in full.
VerdictKindSaidClaimTopic
Not checkable Not checkable 14 Nov 2019 I think each of those criteria are steps towards a G.I. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 the illusory progress gap. You see a little bit of progress on some piece of a problem. And you think that that means we've solved the… Not checkable · 14 Nov 2019
Unresolvable Factual 14 Nov 2019 Most of the current best AI systems fundamentally just learn contingencies rather than general abstract concepts that humans can apply quickly. Factual · 14 Nov 2019 AI capabilities
Not checkable Not checkable 14 Nov 2019 Well, it's idiotic to use a system that is 75 percent right. When you've got a calculator that's 100 percent right. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 Our brains didn't evolve to say, hey, that person's intelligent and that's really just a machine and just get suckered in. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 There isn't a one size fits all, as you're saying, and there's a long tradition in intellectual history of people kind of overreaching in… Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 Of course, Google doesn't understand natural language. It understands a small fragment of natural language, but it's very easy to fool it Not checkable · 14 Nov 2019
Not researched Factual 14 Nov 2019 There is no formal argument out there that the person Marcus is discussing (referred to only as "he") is wrong. Factual · 14 Nov 2019 AI capabilities
Not checkable Not checkable 14 Nov 2019 So you start with systems that are total blank slates. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 we don't even have the right number of miles to even begin to think that we've achieved reliability. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 Just because you see the machine and think that that’s what is doing doesn’t mean that that’s what the machines actually doing. Not checkable · 14 Nov 2019
Not researched Factual 14 Nov 2019 As of November 2019, there was no systematic way of solving this problem and its many variations, and Marcus said he thinks there may be no possibility of doing so. Factual · 14 Nov 2019 AI safety and alignment
Not checkable Not checkable 14 Nov 2019 I think is 100 percent useful to know about them. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 deep learning in particular doesn't really solve. So it evades it by having large amounts of statistics. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 they taught us to speak truth to power at Hampshire College, and that's what Lee and I have in common. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 But there’s there’s a huge bias in how these things are are reported. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 And so a lot of the knowledge is kind of undifferentiated in a way that makes it useless. Not checkable · 14 Nov 2019
Not researched Factual 14 Nov 2019 No current AI system, deep learning or otherwise, can induce heuristics that transfer from one game to the next. Factual · 14 Nov 2019 AI capabilities
Not checkable Not checkable 14 Nov 2019 And that's just not good enough for what we call level 5 autonomy. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 No free lunch there, which basically states that there is no best way to solve any problem or to search any database Not checkable · 14 Nov 2019
Not researched Factual 14 Nov 2019 Nobody knows how to build a machine that can tidy a room containing arbitrary objects. Factual · 14 Nov 2019 Technology
Not checkable Not checkable 14 Nov 2019 You build a robot to go into the real world and things just don’t work that well. Not checkable · 14 Nov 2019
Unresolvable Factual 14 Nov 2019 The best-selling domestic robot of all time is still the Roomba, which does not manipulate objects. Factual · 14 Nov 2019 Technology
Not checkable Not checkable 14 Nov 2019 I think that in a way, they’re kind of tying their hands behind their back, taking away one of the most valuable discoveries in the history… Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 I think it’s a five hundred million dollar project and nobody who has the appetite for it right now. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 I say such things are an empirical question. Let's see how they work in specific domains. Not checkable · 14 Nov 2019
Mixed Factual 14 Nov 2019 AI currently lacks a way of building reusable skills. Factual · 14 Nov 2019 AI capabilities
Not checkable Not checkable 14 Nov 2019 So it is hard to define understanding. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 even though it knows the statistics of what's going on, it doesn't actually understand what's going on. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 It’s a really endemic problem and people think it’s easy to solve and it’s not. Not checkable · 14 Nov 2019
Not researched Factual 14 Nov 2019 Part of the Tesla Model 3 production problem was that Elon Musk overestimated how easy it would be to get robots to work in complex assembly lines. Factual · 14 Nov 2019 AI, the economy and work
Not researched Factual 14 Nov 2019 In the early 1990s nobody had any idea how big the internet would become. Factual · 14 Nov 2019 AI timelines and forecasting
Not checkable Not checkable 14 Nov 2019 science doesn't progress by people being chummy with each other. It progresses by recognizing limitations and taking the next step. Not checkable · 14 Nov 2019
Unresolvable Prediction 14 Nov 2019 The real action in AI will be in finding theoretically sound ways of combining symbolic and deep learning traditions. Prediction · 14 Nov 2019 AI capabilities
Resolved true Factual 14 Nov 2019 AlphaGo is a hybrid system that uses deep learning for pattern recognition and Monte Carlo tree search, a symbolic operation, to search possibilities. Factual · 14 Nov 2019 AI capabilities
Not checkable Not checkable 14 Nov 2019 it's been very hard for me to get the field to engage in that. Not checkable · 14 Nov 2019
Not checkable Not checkable 14 Nov 2019 they might be true insofar as they go, but in a way they go too far. Not checkable · 14 Nov 2019
Not researched Factual 14 Nov 2019 Every neural network Marcus knows of has output nodes that are in fact symbols. Factual · 14 Nov 2019 AI capabilities
Not checkable Not checkable 14 Nov 2019 knowing something about what people do in the process of learning these higher order skills might be very helpful towards A.I. Not checkable · 14 Nov 2019
Not checkable Not checkable 20 Sep 2019 A lot of my recent work is about how we can best use an understanding of the human mind to make AI better in order to make the work a… Not checkable · 20 Sep 2019
Unresolvable Prediction 20 Sep 2019 Marcus said he thinks it is just a question of whether "we will get there" gradually or through a lot of social unrest. Prediction · 20 Sep 2019 AI, the economy and work
Not checkable Not checkable 20 Sep 2019 It’s not one thing. Not checkable · 20 Sep 2019
Not checkable Not checkable 20 Sep 2019 They don’t really understand the world around them. Not checkable · 20 Sep 2019
Resolved true Factual 20 Sep 2019 Driverless cars that are not actually reliable can be easily fooled. Factual · 20 Sep 2019 AI capabilities
Not checkable Not checkable 20 Sep 2019 We just haven’t done enough work on that yet because it’s seductive. Not checkable · 20 Sep 2019
Unresolvable Factual 20 Sep 2019 If the paddle in the game Breakout is moved by three pixels, a deep-learning system trained on the game fails. Factual · 20 Sep 2019 AI capabilities
Not checkable Not checkable 20 Sep 2019 And that entails really thinking about AI pretty differently than how people are doing it now. Not checkable · 20 Sep 2019
Not yet due Prediction 20 Sep 2019 Machines will achieve world-class scientific reasoning within the next 500 years (by 2519). Prediction · 20 Sep 2019 AI timelines and forecasting
Not checkable Not checkable 20 Sep 2019 Evolutionary happenstance has left us with a really lousy memory system. Not checkable · 20 Sep 2019
Unresolvable Prediction 20 Sep 2019 Smarter AI could solve many medical and scientific problems. Prediction · 20 Sep 2019 AI capabilities
Not checkable Not checkable 20 Sep 2019 I don’t think we can go backward and stop trusting machines. Not checkable · 20 Sep 2019
Unresolvable Prediction 20 Sep 2019 The ultimate AI systems will have the memory capabilities of machines rather than of people. Prediction · 20 Sep 2019 AI capabilities
Not checkable Not checkable 20 Sep 2019 Paint your doorknob black, against a black background, which will greatly reduce the chance that the robot will even be able to see it. Not checkable · 20 Sep 2019
Not checkable Not checkable 20 Sep 2019 But we want them to be as flexible as people. Not checkable · 20 Sep 2019
Unresolvable Prediction 20 Sep 2019 Marcus said he expects machines will eventually be able to do world-class scientific reasoning, though they are nowhere near that now. Prediction · 20 Sep 2019 AI timelines and forecasting
Mostly true Factual 20 Sep 2019 Biological cognition involves many different mechanisms, working on different principles, that reinforce and counter-balance each other. Factual · 20 Sep 2019 Science and research methods
Not checkable Not checkable 20 Sep 2019 We’re going to have to stop relying on these first-order approximations and get to things that are causal and more sophisticated. Not checkable · 20 Sep 2019
Mostly true Factual 20 Sep 2019 Deep learning does more than linear regression but is a more sophisticated form of statistical correlation. Factual · 20 Sep 2019 AI capabilities
Not checkable Not checkable 20 Sep 2019 Close your doors, and for good measure, lock them. Not checkable · 20 Sep 2019
Not checkable Not checkable 20 Sep 2019 The negative thing that I’m trying to avoid and help the world to avoid is things like driverless cars that are not actually reliable and… Not checkable · 20 Sep 2019
Not checkable Not checkable 20 Sep 2019 Evolutionary psychology is the field that most asks the question why we do the things that we do, Not checkable · 20 Sep 2019
Not checkable Not checkable 20 Sep 2019 The more intelligence you have, the more you can be robust, Not checkable · 20 Sep 2019
Mostly true Factual 20 Sep 2019 Bad AI is leading to fatalities and undermining democracy. Factual · 20 Sep 2019 AI policy and governance
Not checkable Not checkable 20 Sep 2019 I don’t think it’s going to work that way. Not checkable · 20 Sep 2019
Resolved true Prediction 20 Sep 2019 Marcus said he did not think machines would reach world-class scientific reasoning within the next five years (by 2024). Prediction · 20 Sep 2019 AI timelines and forecasting
Unresolvable Prediction 20 Sep 2019 Marcus said he thinks it is inevitable that society will adopt universal basic income, and that the open question is whether this happens gradually or through social unrest. Prediction · 20 Sep 2019 AI, the economy and work
Not checkable Not checkable 20 Sep 2019 I’m a fan of Judea Pearl’s recent book, The Book of Why, which is about causality, Not checkable · 20 Sep 2019
Not checkable Not checkable 20 Sep 2019 It’s not that we want to make airplanes that fly exactly like birds. Not checkable · 20 Sep 2019
Not checkable Not checkable 20 Sep 2019 I think that we are entering an era where we are trusting machines more and more, but the machines don’t yet deserve that trust—they… Not checkable · 20 Sep 2019
Not checkable Not checkable 20 Sep 2019 It’s really seductive to use these statistical calculators to get a good first approximation to a lot of things. Not checkable · 20 Sep 2019
Not checkable Not checkable 20 Sep 2019 The thing I find most inspiring is nature itself, Not checkable · 20 Sep 2019
Not checkable Not checkable 20 Sep 2019 People are looking for a silver bullet, Not checkable · 20 Sep 2019
Not checkable Not checkable 20 Sep 2019 Causality is really important to genuine intelligence. Not checkable · 20 Sep 2019
Not checkable Not checkable 20 Sep 2019 It’s mostly about statistics, Not checkable · 20 Sep 2019
Not checkable Not checkable 20 Sep 2019 Our view is that stuff is never robust. Not checkable · 20 Sep 2019
Not checkable Not checkable 31 Mar 2019 Scary and important work on AI and cybersecurity: Not checkable · 31 Mar 2019
Not checkable Not checkable 31 Mar 2019 shows Tesla system lacks adequate redundancy and common sense. Not checkable · 31 Mar 2019
Not checkable Not checkable 28 Mar 2019 For anyone who thought I was jousting at strawmen Not checkable · 28 Mar 2019
Not checkable Not checkable 6 Mar 2019 Blockheads! fun new book on Ned Block and philosophy of mind & consciousness Not checkable · 6 Mar 2019
Not checkable Not checkable 18 Feb 2019 this is great! Not checkable · 18 Feb 2019
Not checkable Not checkable 18 Feb 2019 Important and thought-provoking technical thread/paper on Imagenet and limits within current deep learning models Not checkable · 18 Feb 2019
Not checkable Not checkable 17 Feb 2019 Retweet this if you care about our country. Not checkable · 17 Feb 2019
Not checkable Not checkable 16 Feb 2019 completely agree: remarkable ≠ semantically coherent. Not checkable · 16 Feb 2019
Not checkable Not checkable 16 Feb 2019 risks of trusting ML too much in biomedicine Not checkable · 16 Feb 2019
Resolved true Factual 16 Feb 2019 The system has no idea that a unicorn has one horn and asserts that it has four. Factual · 16 Feb 2019 AI capabilities
Unresolvable Factual 16 Feb 2019 OpenAI's new system is state of the art on some question-answering tests but is no match for humans on story understanding. Factual · 16 Feb 2019 AI capabilities
Not checkable Not checkable 16 Feb 2019 some important hints into what’s going on in that new OpenAi language model Not checkable · 16 Feb 2019
Not checkable Not checkable 6 Feb 2019 history doesn’t repeat itself - except when it does. Not checkable · 6 Feb 2019
Not checkable Not checkable 3 Feb 2019 for the foreseaable future our biggest fear should not be AI rising up. Not checkable · 3 Feb 2019
Not checkable Not checkable 3 Feb 2019 it should be AI (& software engineering in general) falling down on the job. Not checkable · 3 Feb 2019
Not researched Factual 31 Jan 2019 Facebook's stock price rose 10% on 2019-01-31. Factual · 31 Jan 2019 Economics and markets
Not checkable Not checkable 30 Jan 2019 This thread is well-worth reading, to understand the brazeness of Facebook, in technical detail. Not checkable · 30 Jan 2019
Not checkable Not checkable 30 Jan 2019 deepmind’s latest publicity stunt, dissected. Not checkable · 30 Jan 2019
Not checkable Not checkable 30 Jan 2019 Spying on teenagers is not totally cool and it certainly shouldn’t be totally legal. Not checkable · 30 Jan 2019
Not checkable Not checkable 25 Jan 2019 Sanders just wrote her own epigraph, right next to Nixon’s “I am not a crook”. Not checkable · 25 Jan 2019
Not checkable Not checkable 25 Jan 2019 this is what she will be remembered for; another career destroyed. Not checkable · 25 Jan 2019
Not checkable Not checkable 21 Jan 2019 This is the most damning paragraph about a major political figure’s loyalties that I have ever read. Not checkable · 21 Jan 2019
Not checkable Not checkable 18 Jan 2019 Handy tool for analyzing utterances from POTUS and his circle. Not checkable · 18 Jan 2019
Not checkable Not checkable 18 Jan 2019 enough already. impeach the liar in chief. Not checkable · 18 Jan 2019
Mostly false Factual 17 Jan 2019 According to Giuliani's disclosure on 2019-01-17, every tweet and statement like the one referenced was a lie, and Trump's campaign colluded. Factual · 17 Jan 2019 Other
Back to the list

Gary Marcus

Open its page

Factual AI capabilities

“So every neural network that I know has, for example, output notes that are in fact symbols.”

Gary Marcus, podcast guest 14 November 2019 EP25 Gary Marcus on Rebooting AI (The Jim Rutt Show)

Every neural network Marcus knows of has output nodes that are in fact symbols.

Not researched Checkable, but outside the sample that was researched.

Cite this claim

Gary Marcus, “Every neural network Marcus knows of has output nodes that are in fact symbols.” (said 14 November 2019). Verdict: not researched. Reliability API, claim c_fda862745f15, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_fda862745f15

Claims without a verdict

Is a verdict here wrong, or do you speak for Gary Marcus? Reply on the claim's own page, with the form at its foot (how replies work). Verdicts are made automatically; how verdicts and scores are made.