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

42 claims

42 claims, sorted by newest said. Choose a claim to read it in full.
VerdictKindSaidClaimTopic
Resolved true Factual 25 Sep 2026 OpenAI's software attacked Australian government servers. Factual · 25 Sep 2026 AI safety and alignment
Resolved true Prediction 27 Jul 2025 OpenAI will release GPT-5 imminently. Prediction · 27 Jul 2025 AI capabilities
Resolved true Prediction 27 Jul 2025 GPT-5 will still make a significant number of obvious errors that are hard to predict. Prediction · 27 Jul 2025 AI capabilities
Resolved true Prediction 27 Jul 2025 If the model being discussed has guardrails like ChatGPT's, they will teeter between being too weak (beaten by jailbreaks) and too strong (rejecting some perfectly reasonable requests). Prediction · 27 Jul 2025 AI safety and alignment
Resolved true Prediction 27 Jul 2025 GPT-5 by itself will not be a general-purpose AGI capable of arbitrary tasks. Prediction · 27 Jul 2025 AI timelines and forecasting
Resolved true Prediction 1 Jan 2025 AI Safety Institutes will offer guidance in 2025 but have little legal authority to stop truly dangerous models. Prediction · 1 Jan 2025 AI policy and governance
Resolved true Factual 1 Jan 2025 As of end 2024, AI model makers have very little moat, with many similar models and a price war. Factual · 1 Jan 2025 Economics and markets
Resolved true Prediction 1 Jan 2025 Profits from AI models will remain modest or nonexistent in 2025. Prediction · 1 Jan 2025 Economics and markets
Resolved true Prediction 1 Jan 2025 Lack of reliability will continue to afflict generative AI in 2025. Prediction · 1 Jan 2025 AI capabilities
Resolved true Prediction 1 Jan 2025 Artificial general intelligence will not be achieved in 2025. Prediction · 1 Jan 2025 AI timelines and forecasting
Resolved true Prediction 1 Jan 2025 Hallucinations will continue to afflict generative AI in 2025. Prediction · 1 Jan 2025 AI capabilities
Resolved true Prediction 1 Jan 2025 Less than 10% of the workforce will be replaced by AI in 2025, probably less than 5%. Prediction · 1 Jan 2025 AI, the economy and work
Resolved true Prediction 1 Jan 2025 The US will continue to have very little regulation protecting consumers from generative AI risks in 2025. Prediction · 1 Jan 2025 AI policy and governance
Resolved true Prediction 1 Jan 2025 AI agents will be heavily hyped throughout 2025 but remain far from reliable, except possibly in very narrow use cases. Prediction · 1 Jan 2025 AI capabilities
Resolved true Prediction 1 Jan 2025 AI power consumption will rise in 2025 and remain a major problem, while few generative AI companies will be transparent about their usage. Prediction · 1 Jan 2025 AI compute, hardware and data
Resolved true Prediction 1 Jan 2025 No single AI system will solve more than 4 of the AI 2027 Marcus-Brundage tasks by the end of 2025. Prediction · 1 Jan 2025 AI capabilities
Resolved true Prediction 1 Jan 2025 In 2025 humanoid robotics will see a lot of hype, but nobody will release a robot remotely as capable as Rosie the Robot. Prediction · 1 Jan 2025 AI capabilities
Resolved true Factual 2025 Bolting extra components onto an LLM and scaling it up has so far not produced human-level AGI. Factual · 2025 AI capabilities
Resolved true Factual 9 Nov 2024 Gary Marcus has for years warned that AI scaling by adding data and compute without architectural changes would not continue forever. Factual · 9 Nov 2024 Their own work or organisation
Resolved true Factual Feb 2024 After Sora's February 2024 announcement, Marcus predicted its physics troubles would persist, and they have. Factual · Feb 2024 Their own work or organisation
Resolved true Factual 26 Oct 2023 Bengio et al.'s proposal states that research breakthroughs are needed to solve some of today's technical challenges in creating AI with safe and ethical objectives. Factual · 26 Oct 2023 AI safety and alignment
Resolved true Factual 26 Oct 2023 There is currently no way to guarantee that large language models won't be jailbroken into saying unethical things. Factual · 26 Oct 2023 AI safety and alignment
Resolved true Factual 30 Mar 2023 The FLI open letter called for more AI research. Factual · 30 Mar 2023 Their own work or organisation
Delivered on time Promise 30 Mar 2023 Gary Marcus will speak at TED about short-term and long-term AI risk and what to do about it, in about two weeks from 2023-03-30. Promise · 30 Mar 2023 Their own work or organisation
Resolved true Factual 30 Mar 2023 The FLI open letter did not call for a ban on AI or for a permanent ban. Factual · 30 Mar 2023 Their own work or organisation
Resolved true Factual 24 Mar 2023 There have already been reports of contamination in GPT-4's training data. Factual · 24 Mar 2023 AI capabilities
Resolved true Factual 24 Mar 2023 GPT-4 has not solved hallucinations, reliability or planning on complex tasks, as the paper's authors acknowledge for planning. Factual · 24 Mar 2023 AI capabilities
Resolved true Prediction 25 Dec 2022 GPT-4 will definitely seem smarter than its predecessors, but its internal architecture will remain problematic. Prediction · 25 Dec 2022 AI capabilities
Resolved true Factual Mar 2022 In March 2022 Gary Marcus published an article arguing that deep learning is hitting a wall. Factual · Mar 2022 Their own work or organisation
Resolved true Factual 16 Nov 2021 In open-ended tasks such as driving, humans outperform AI systems. Factual · 16 Nov 2021 AI capabilities
Resolved true Factual 1 Nov 2021 Aaron Beck, developer of cognitive therapy, has died. Factual · 1 Nov 2021 Other
Resolved true Factual 17 Apr 2021 Garry Kasparov beat Gary Marcus at a chess exhibition. Factual · 17 Apr 2021 Other
Resolved true Factual 13 Apr 2021 Gary Marcus published a 1999 Science article on infant rule learning. Factual · 13 Apr 2021 Science and research methods
Resolved true Factual 12 May 2020 When the environment changes, as during the pandemic, many AI models start to break. Factual · 12 May 2020 AI capabilities
Resolved true Factual 30 Apr 2020 A piece on common sense in AI by John Pavlus includes some chortling and skepticism from Gary Marcus. Factual · 30 Apr 2020 AI capabilities
Resolved true Factual 14 Nov 2019 Gary Marcus wrote a book called The Birth of the Mind about how a small number of genes builds a complex brain. Factual · 14 Nov 2019 Their own work or organisation
Resolved true Factual 14 Nov 2019 Cyc is not an end-to-end system. Factual · 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
Resolved true Factual 20 Sep 2019 Driverless cars that are not actually reliable can be easily fooled. Factual · 20 Sep 2019 AI capabilities
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
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
Resolved true Factual 2 Jan 2019 Gary Marcus posted "Deep Learning: A Critical Appraisal" on 2018-01-02, asking whether deep learning might be approaching a wall and be just one tool among many needed for AGI. Factual · 2 Jan 2019 AI capabilities
Back to the list

Gary Marcus

Open its page

Factual AI safety and alignment

“We have no way of guaranteeing that large language models won’t be jailbroken into saying unethical things.”

Gary Marcus, author of Marcus on AI 26 October 2023 A CERN for AI and the Global Governance of AI (Marcus on AI)

There is currently no way to guarantee that large language models won't be jailbroken into saying unethical things.

Resolved true The record shows it happened or was so.

What would count as true or false

Judged as of 2023-10-26. True: the evidence from that time shows that no technique existed that provably or reliably stopped LLMs from being jailbroken into producing harmful or unethical output, and that jailbreaks of leading models were being shown routinely (e.g. the universal adversarial suffixes reported in 2023). Mostly true: the core claim holds, with only minor caveats such as narrow formal guarantees for small or restricted settings. Mixed: some methods offered strong but not complete protection, so whether a 'guarantee' existed was genuinely disputed. Mostly false: robust defences were broadly accepted as effective, and the claim overstates the remaining gaps. False: a method that guarantees protection against jailbreaks was established and accepted by then.

Evidence

  1. Researchers Discover New Vulnerability in Large Language Models Carnegie Mellon University (News) · 28 July 2023
    ““Right now, we simply don’t have a convincing way to stop this from happening, so the next step is to figure out how to fix these models,” Fredrikson said.”
  2. Researchers Discover New Vulnerability in Large Language Models Carnegie Mellon University (News) · 28 July 2023
    “By training the attack suffix on multiple prompts and models, the researchers have also induced objectionable content in public interfaces like Google Bard and Claud and in open-source LLMs such as Llama 2 Chat, Pythia, Falcon and others.”
  3. Jailbroken: How Does LLM Safety Training Fail? arXiv (authors from academic institutions) · 13 April 2023
    “We find that vulnerabilities persist despite the extensive red-teaming and safety-training efforts behind these models. Notably, new attacks utilizing our failure modes succeed on every prompt in a collection of unsafe requests from the models’ red-teaming evaluation sets”
  4. AI Safety Newsletter #17 Center for AI Safety · 1 August 2023
    “and the companies have taken preliminary measures to mitigate the attacks. But no robust defenses are currently known, necessitating two key responses to the threat of adversarial attacks.”

Reasoning

In July 2023, CMU researchers reported universal adversarial suffixes that drew objectionable content from Bard, Claude, Llama 2 Chat and other models. One of them said there was no convincing way to stop such attacks (E1, E2). An academic paper found that jailbreaks still worked despite extensive red-teaming and safety training (E3). The Center for AI Safety wrote in August 2023 that no robust defences were known (E4). Together, these independent sources show that as of October 2023 no method could guarantee protection against jailbreaks, and that leading models were being jailbroken routinely.

Cite this claim

Gary Marcus, “There is currently no way to guarantee that large language models won't be jailbroken into saying unethical things.” (said 26 October 2023). Verdict: resolved true. Reliability API, claim c_eeee4d06cd37, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_eeee4d06cd37

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.