Claims

2,653 claims by 33 public figures and organisations, checked against the public record. Each has its quote, source, verdict and evidence. Read them without a key; browse by person.

28,548 claims, sorted by newest said. Choose a claim to read it in full.
VerdictPersonKindSaidClaimTopic
Own records only Dario Amodei Factual 8 Aug 2023 Anthropic is focusing heavily on cybersecurity, including working with its cloud providers. Dario Amodei · Factual · 8 Aug 2023 Their own work or organisation
Not researched Dario Amodei Factual 8 Aug 2023 Anthropic works with cloud providers for its data centers. Dario Amodei · Factual · 8 Aug 2023 Their own work or organisation
Unresolvable Dario Amodei Prediction 8 Aug 2023 Serious danger from scaled AI models will first come from misuse, such as biorisk, rather than misalignment. Dario Amodei · Prediction · 8 Aug 2023 AI safety and alignment
Not checkable Dario Amodei Not checkable 8 Aug 2023 I actually think fact number one and fact number two are enough to be really worried. Dario Amodei · Not checkable · 8 Aug 2023
Not checkable Dario Amodei Not checkable 8 Aug 2023 I think that's a mix of just the model skill level is not high enough yet. Dario Amodei · Not checkable · 8 Aug 2023
Not checkable Dario Amodei Not checkable 8 Aug 2023 That's not my position at all. Dario Amodei · Not checkable · 8 Aug 2023
Not checkable Dario Amodei Not checkable 8 Aug 2023 Mechanistic interpretability is the only thing that even in principle is the thing where it's more like an X-ray of the model than… Dario Amodei · Not checkable · 8 Aug 2023
Own records only Dario Amodei Factual 8 Aug 2023 Every traditional investor who invests in Anthropic looks at its unusual governance body. Dario Amodei · Factual · 8 Aug 2023 Their own work or organisation
Not checkable Dario Amodei Not checkable 8 Aug 2023 It feels to me like an intern in most areas, but then some specific areas where it's better than that. Dario Amodei · Not checkable · 8 Aug 2023
Unresolvable Dario Amodei Prediction 8 Aug 2023 Dario Amodei said he personally thinks it is very unlikely that AI scaling laws will just stop. Dario Amodei · Prediction · 8 Aug 2023 AI capabilities
Not checkable Dario Amodei Not checkable 8 Aug 2023 I think then you would have to go for some kind of RL. Dario Amodei · Not checkable · 8 Aug 2023
Not checkable Dario Amodei Not checkable 8 Aug 2023 I thought it had understood a lot of the essence of language but then there's further to go. Dario Amodei · Not checkable · 8 Aug 2023
Not checkable Dario Amodei Not checkable 8 Aug 2023 I guess if you had the view that alignment is completely unsolvable, then you'd be like, well, we're dead anyway so I don't want to worry… Dario Amodei · Not checkable · 8 Aug 2023
Not researched Dario Amodei Factual 8 Aug 2023 Email accounts of fairly high-ranking US government officials were recently hacked via Microsoft, which provided the accounts. Dario Amodei · Factual · 8 Aug 2023 Other global risks
Not researched Dario Amodei Factual 8 Aug 2023 As of 2023, AI models are only beginning to prove relatively simple mathematical theorems. Dario Amodei · Factual · 8 Aug 2023 AI capabilities
Not researched Ray Kurzweil Factual Aug 2023 Large language models operate thousands of times faster than humans. Ray Kurzweil · Factual · Aug 2023 AI capabilities
Resolved true Ray Kurzweil Factual Aug 2023 In The Age of Spiritual Machines (1999), Kurzweil predicted computers would pass the Turing Test by 2029. Ray Kurzweil · Factual · Aug 2023 AI timelines and forecasting
Not researched Ray Kurzweil Factual Aug 2023 The system he was describing does not get a particular task involving a sentence put in quotes correct, though humans can do it easily. Ray Kurzweil · Factual · Aug 2023 AI capabilities
Not researched Ray Kurzweil Factual Aug 2023 The consensus of AI experts around the world is now also that the Turing Test will be passed by 2029. Ray Kurzweil · Factual · Aug 2023 AI timelines and forecasting
Not checkable Jeremy Grantham Not checkable 5 Jul 2023 I’m only interested in the really great bubbles like 1929, 2000, and 2021, [which] are the three senior bubbles in [the] U.S stock market. Jeremy Grantham · Not checkable · 5 Jul 2023
Not checkable Jeremy Grantham Not checkable 5 Jul 2023 We can do that before the real effects of A.I. kick in. Jeremy Grantham · Not checkable · 5 Jul 2023
Unresolvable Jeremy Grantham Prediction 5 Jul 2023 Corporate profit margins have already fallen a decent amount but could fall considerably further. Jeremy Grantham · Prediction · 5 Jul 2023 Economics and markets
Not checkable Jeremy Grantham Not checkable 5 Jul 2023 It’s very difficult to tell. Jeremy Grantham · Not checkable · 5 Jul 2023
Not checkable Jeremy Grantham Not checkable 5 Jul 2023 Lord knows this was complicated before A.I. raised its ugly head. We had inflation, the Fed, how quickly would rates go up, how far would… Jeremy Grantham · Not checkable · 5 Jul 2023
Unresolvable Jeremy Grantham Prediction 5 Jul 2023 Grantham's guess was that within a year or two of mid-2023 there would be time for a fairly traditional bubble losing air, a fairly traditional recession, a fairly traditional decline in profit margins and some grief in the stock markets. Jeremy Grantham · Prediction · 5 Jul 2023 Economics and markets
Not checkable Jeremy Grantham Not checkable 5 Jul 2023 And how badly other economic variables will be—the trouble within global trade, the trouble with China, the trouble with the war. Jeremy Grantham · Not checkable · 5 Jul 2023
Not checkable Jeremy Grantham Not checkable 5 Jul 2023 Some of the brightest people on the planet say it’s all nonsense, it’s just a parrot learning by trial and error. Jeremy Grantham · Not checkable · 5 Jul 2023
Not checkable Jeremy Grantham Not checkable 5 Jul 2023 We have checked off pretty well every one of the boxes, Jeremy Grantham · Not checkable · 5 Jul 2023
Not checkable Jeremy Grantham Not checkable 5 Jul 2023 Other people say it will change everything, it will double productivity and everything in between, Jeremy Grantham · Not checkable · 5 Jul 2023
Mostly true Jeremy Grantham Factual 5 Jul 2023 Since ChatGPT and October-November 2022, there has been a new little flurry of market interest in AI that is very concentrated. Jeremy Grantham · Factual · 5 Jul 2023 Economics and markets
Not checkable Marc Andreessen Not checkable 29 Jun 2023 By the way, in some ways it actually has interesting psychological similarities to some of the COVID hysteria. So there's just a lot of… Marc Andreessen · Not checkable · 29 Jun 2023
Not checkable Marc Andreessen Not checkable 29 Jun 2023 From my perspective it's this very kind of a-technical, a-scientific, kind of hysterical thing that's playing out, Marc Andreessen · Not checkable · 29 Jun 2023
Not checkable Marc Andreessen Not checkable 29 Jun 2023 The hysterics have a much bigger voice than people who are calmer and maybe a little bit more dispassionate about things. Marc Andreessen · Not checkable · 29 Jun 2023
Not checkable Marc Andreessen Not checkable 29 Jun 2023 The hysterical freakout has arrived in Washington, and it's arrived in Brussels, and it's arriving in capitals of countries all over the… Marc Andreessen · Not checkable · 29 Jun 2023
Not checkable Marc Andreessen Not checkable 29 Jun 2023 The demands for regulation and laws under conditions of hysteria have kind of reached the point where I felt like I couldn't just let that… Marc Andreessen · Not checkable · 29 Jun 2023
Unresolvable Yann LeCun Factual 28 Jun 2023 MIT Technology Review published a piece by Melissa Heikkilä about Meta's stance on AI existential risks, with quotes from Joelle Pineau and Yann LeCun. Yann LeCun · Factual · 28 Jun 2023 AI safety and alignment
Not researched Yann LeCun Factual 28 Jun 2023 Data center power consumption barely changes because it is largely determined by economics: Google and Meta cannot spend more than a few dollars per user per year on power. Yann LeCun · Factual · 28 Jun 2023 Economics and markets
Mostly true Yann LeCun Factual 28 Jun 2023 The US is a clear outlier among developed countries in healthcare cost and outcomes. Yann LeCun · Factual · 28 Jun 2023 Other
Mostly true Yann LeCun Factual 28 Jun 2023 The audience YES/NO split at the Munk Debate was 67/33 before the debate and 63/37 after. Yann LeCun · Factual · 28 Jun 2023 AI safety and alignment
Mostly true Yann LeCun Factual 28 Jun 2023 In the official result of the Munk Debate on AI existential threat, the NO side (Mitchell and LeCun) won by 4% over the YES side (Tegmark and Bengio). Yann LeCun · Factual · 28 Jun 2023 AI safety and alignment
Not checkable Rodney Brooks Not checkable 28 Jun 2023 Such an unexpected honor. Rodney Brooks · Not checkable · 28 Jun 2023
Not checkable Yann LeCun Not checkable 28 Jun 2023 Also, my ears are not pointy, and my archer skills are feeble. Yann LeCun · Not checkable · 28 Jun 2023
Not checkable Yann LeCun Not checkable 27 Jun 2023 Nice piece. But it should have talked about @NablaTech one of the first players in that space (which the NYT wrote about a while back). Yann LeCun · Not checkable · 27 Jun 2023
Resolved true Yann LeCun Factual 27 Jun 2023 Clément Delangue defended AI open science and open source in his congressional testimony. Yann LeCun · Factual · 27 Jun 2023 AI policy and governance
Not checkable Yann LeCun Not checkable 26 Jun 2023 We are missing a number of basic concepts to get human-level AI. Yann LeCun · Not checkable · 26 Jun 2023
Not checkable Yann LeCun Not checkable 26 Jun 2023 I also agree that many of the AI safety policies that have been floating around are naïve. Yann LeCun · Not checkable · 26 Jun 2023
Not checkable Yann LeCun Not checkable 26 Jun 2023 Good thread on AI policy from @jackclarkSF Yann LeCun · Not checkable · 26 Jun 2023
Not checkable Yann LeCun Not checkable 26 Jun 2023 I agree that limiting compute or requiring a license above a certain threshold of compute power would be bad policy. Yann LeCun · Not checkable · 26 Jun 2023
Not checkable Yann LeCun Not checkable 25 Jun 2023 Thanks for clarifying. Yann LeCun · Not checkable · 25 Jun 2023
Not checkable Yann LeCun Not checkable 25 Jun 2023 On le construira pour être bénéfique. Yann LeCun · Not checkable · 25 Jun 2023
Resolved true Rodney Brooks Factual 24 Jun 2023 Rodney Brooks gave a keynote at CVPR 2023 on Tuesday 20 June 2023. Rodney Brooks · Factual · 24 Jun 2023 Their own work or organisation
Not checkable Yann LeCun Not checkable 23 Jun 2023 He doesn't seem worried about the probability that he will produce intelligent evil psychopaths. Yann LeCun · Not checkable · 23 Jun 2023
Resolved true Yann LeCun Factual 23 Jun 2023 Max Tegmark is a professor at MIT. Yann LeCun · Factual · 23 Jun 2023 Their own work or organisation
Not checkable Yann LeCun Not checkable 23 Jun 2023 We can do the same with AI: we can thoroughly test their benevolence before we give them any knowledge and agency. Yann LeCun · Not checkable · 23 Jun 2023
Not researched Yann LeCun Factual 22 Jun 2023 A large number of Americans believe in conspiracy theories, though the proportion has not increased over time. Yann LeCun · Factual · 22 Jun 2023 Philosophy, ethics and society
Not researched Yann LeCun Factual 22 Jun 2023 Yann LeCun gave an opening statement at the Munk Debate on the existential risk of AI on 2023-06-22. Yann LeCun · Factual · 22 Jun 2023 AI safety and alignment
Own records only Rodney Brooks Factual 22 Jun 2023 A Waymo commercial was being filmed across the street from a pickup location in San Francisco in June 2023. Rodney Brooks · Factual · 22 Jun 2023 Technology
Resolved true Rodney Brooks Factual 22 Jun 2023 The San Francisco fire chief is unhappy with driverless vehicles in the city. Rodney Brooks · Factual · 22 Jun 2023 Technology
Not checkable Yann LeCun Not checkable 22 Jun 2023 Automated manufacturing makes handcrafted goods more valuable. Yann LeCun · Not checkable · 22 Jun 2023
Not checkable Rodney Brooks Not checkable 22 Jun 2023 Moving fast and breaking things. Rodney Brooks · Not checkable · 22 Jun 2023
Unresolvable Rodney Brooks Prediction 21 Jun 2023 Humanity will survive large language models. Rodney Brooks · Prediction · 21 Jun 2023 AI safety and alignment
Not checkable Yann LeCun Not checkable 21 Jun 2023 Fun CIFAR workshop, as usual. Yann LeCun · Not checkable · 21 Jun 2023
Unresolvable Yann LeCun Prediction 20 Jun 2023 Objective-Driven AI architectures are the future of AI. Yann LeCun · Prediction · 20 Jun 2023 AI capabilities
Unresolvable Yann LeCun Factual 19 Jun 2023 A joint interview of Yoshua Bengio and Yann LeCun was published in Les Échos Weekend a few days before 2023-06-19. Yann LeCun · Factual · 19 Jun 2023 Their own work or organisation
Not checkable Yann LeCun Not checkable 19 Jun 2023 I can't disagree! Yann LeCun · Not checkable · 19 Jun 2023
Not researched Yann LeCun Factual 18 Jun 2023 Meta AI / FAIR has a large presence at CVPR 2023. Yann LeCun · Factual · 18 Jun 2023 Their own work or organisation
Not checkable Yann LeCun Not checkable 18 Jun 2023 How to shoot yourself in the foot with a bazooka in the form of insane immigration laws. Yann LeCun · Not checkable · 18 Jun 2023
Not researched Yann LeCun Factual 18 Jun 2023 Yann LeCun is an advisor to Enhance Lab. Yann LeCun · Factual · 18 Jun 2023 Their own work or organisation
Not researched Yann LeCun Factual 15 Jun 2023 Yann LeCun expressed opinions on the future of AI at the Innovation Press Day held at Meta-Paris on Tuesday 2023-06-13. Yann LeCun · Factual · 15 Jun 2023 Their own work or organisation
Not checkable Yann LeCun Not checkable 15 Jun 2023 A good article by the BBC about my views on the future of AI. Yann LeCun · Not checkable · 15 Jun 2023
Not researched Yann LeCun Factual 14 Jun 2023 Ania Dawid transcribed Yann LeCun's lectures on Energy-Based Models, which he gave at the Les Houches summer school in 2022. Yann LeCun · Factual · 14 Jun 2023 Science and research methods
Mostly true Marc Andreessen Factual 12 Jun 2023 The Transformer algorithm, a neural net form that works in a sort of generalized way, was a recent theoretical breakthrough. Marc Andreessen · Factual · 12 Jun 2023 AI capabilities
Mixed Marc Andreessen Factual 12 Jun 2023 AI development has a critical mass basically in two places in the world, the United States and China. Marc Andreessen · Factual · 12 Jun 2023 AI compute, hardware and data
Not researched Marc Andreessen Factual 12 Jun 2023 As of June 2023, AI chatbots' context window is limited and restarts from scratch in a new window. Marc Andreessen · Factual · 12 Jun 2023 AI capabilities
Resolved true Marc Andreessen Factual 12 Jun 2023 The Transformer algorithm was invented in 2017. Marc Andreessen · Factual · 12 Jun 2023 AI capabilities
Unresolvable Marc Andreessen Factual 12 Jun 2023 As of June 2023, AI breakthroughs were happening essentially every week. Marc Andreessen · Factual · 12 Jun 2023 AI capabilities
Unresolvable Marc Andreessen Factual 12 Jun 2023 GPT-4 has far superior crystallized intelligence to any human. Marc Andreessen · Factual · 12 Jun 2023 AI capabilities
Not researched Marc Andreessen Factual 12 Jun 2023 People of higher intelligence show higher physical health, educational attainment, career success and parenting success. Marc Andreessen · Factual · 12 Jun 2023 Philosophy, ethics and society
Not checkable Marc Andreessen Not checkable 12 Jun 2023 69,420 Marc Andreessen · Not checkable · 12 Jun 2023
Unresolvable Marc Andreessen Prediction 12 Jun 2023 Freedom-oriented regimes will generally want to make the US bet on AI. Marc Andreessen · Prediction · 12 Jun 2023 AI policy and governance
Unresolvable Marc Andreessen Prediction 12 Jun 2023 It is entirely possible that fundamentally better neural network architectures will now be invented. Marc Andreessen · Prediction · 12 Jun 2023 AI timelines and forecasting
Unresolvable Marc Andreessen Prediction 12 Jun 2023 Authoritarian-oriented regimes will be tempted to take the Chinese path on AI. Marc Andreessen · Prediction · 12 Jun 2023 AI policy and governance
Unresolvable Marc Andreessen Factual 12 Jun 2023 People of higher intelligence are less violent and more capable of solving problems. Marc Andreessen · Factual · 12 Jun 2023 Philosophy, ethics and society
Unresolvable Marc Andreessen Prediction 12 Jun 2023 Many countries around the world will have to make a fundamental choice between US and Chinese AI. Marc Andreessen · Prediction · 12 Jun 2023 AI policy and governance
Not researched Marc Andreessen Factual 12 Jun 2023 There have been AI booms in the past that did not pay off. Marc Andreessen · Factual · 12 Jun 2023 AI, the economy and work
Not researched Marc Andreessen Factual 12 Jun 2023 Andreessen said that Google "kind of let" a technology that emerged in 2017 "sit on the shelf for several years", and that OpenAI then picked up and implemented it. Marc Andreessen · Factual · 12 Jun 2023 AI, the economy and work
Mostly false Marc Andreessen Factual 12 Jun 2023 Research establishes that in human affairs, higher intelligence leads to better outcomes in essentially every measurable aspect of human welfare and life quality. Marc Andreessen · Factual · 12 Jun 2023 Philosophy, ethics and society
Not researched Marc Andreessen Factual 12 Jun 2023 Within a single session, current AI chatbots build up a context window and retain state across queries. Marc Andreessen · Factual · 12 Jun 2023 AI capabilities
Unresolvable Marc Andreessen Prediction 12 Jun 2023 Which AI bet (US or China) a country takes will help tilt its political system in the future, as AI shapes its politics and culture. Marc Andreessen · Prediction · 12 Jun 2023 AI policy and governance
Not researched Marc Andreessen Factual 12 Jun 2023 Because Germany is averse to nuclear power, it is burning hydrocarbons and funding the Russian invasion of Ukraine through energy purchases. Marc Andreessen · Factual · 12 Jun 2023 Other global risks
Unresolvable Marc Andreessen Prediction 12 Jun 2023 There may be decades of architectural breakthroughs still ahead in AI. Marc Andreessen · Prediction · 12 Jun 2023 AI timelines and forecasting
Not researched Marc Andreessen Factual 12 Jun 2023 Within the United States, AI development is highly concentrated in the San Francisco Bay Area. Marc Andreessen · Factual · 12 Jun 2023 AI compute, hardware and data
Unresolvable Marc Andreessen Factual 12 Jun 2023 Andreessen said he would estimate GPT-4's human-equivalent IQ, which he described as fluid intelligence, at roughly 130 to 135. Marc Andreessen · Factual · 12 Jun 2023 AI capabilities
Mostly true Marc Andreessen Factual 12 Jun 2023 As of June 2023, the U.S. government was both alarmed about deepfakes and trying to outlaw blockchains. Marc Andreessen · Factual · 12 Jun 2023 AI policy and governance
Mixed Marc Andreessen Prediction 12 Jun 2023 Andreessen said he was pretty confident that within roughly two years of June 2023, AI systems' limited persistent memory across sessions would be a solved problem. Marc Andreessen · Prediction · 12 Jun 2023 AI timelines and forecasting
Mostly true Marc Andreessen Factual 12 Jun 2023 As of June 2023 there are chip shortages throughout the AI ecosystem, and NVIDIA cannot make enough chips. Marc Andreessen · Factual · 12 Jun 2023 AI compute, hardware and data
Not researched Marc Andreessen Factual 12 Jun 2023 Querying a system like GPT-4 twice with the same question gives two different answers. Marc Andreessen · Factual · 12 Jun 2023 AI capabilities
Not checkable Marc Andreessen Not checkable 12 Jun 2023 It's time to build. Marc Andreessen · Not checkable · 12 Jun 2023
Not researched Marc Andreessen Factual 12 Jun 2023 There were plenty of scams and fraud before AI. Marc Andreessen · Factual · 12 Jun 2023 Other
Not researched Marc Andreessen Factual 12 Jun 2023 About 99% of AI development is happening in the San Francisco Bay Area. Marc Andreessen · Factual · 12 Jun 2023 AI compute, hardware and data
Back to the list

Yann LeCun

Open its page

Factual AI safety and alignment

“The NO side (@MelMitchell1 and me) wins by 4% over the YES side (@tegmark and Yoshua Bengio).”

Yann LeCun, own X post 28 June 2023 Post by @ylecun on X, 2023-06-28 (X)

In the official result of the Munk Debate on AI existential threat, the NO side (Mitchell and LeCun) won by 4% over the YES side (Tegmark and Bengio).

Mostly true True in substance, wrong in a detail.

What would count as true or false

Judge against the official result the Munk Debates published for the June 2023 debate on whether AI research and development poses an existential threat. TRUE: the official result declares the NO side (Mitchell and LeCun) the winner by 4 percentage points (3.5 to 4.5 counts as 4) on the measure Munk uses to name a winner (normally the gain in audience vote share from the pre-debate poll to the post-debate poll). MOSTLY TRUE: NO is the official winner and the margin on that measure is off by a little more, between about 1.5 and 5.5 points. MOSTLY FALSE: NO is named winner but the margin is clearly not about 4 points, or the 4% figure is real but comes from a measure Munk did not use to pick the winner (e.g., NO lost the final vote but gained 4 points), so the claim misstates what decided the result. FALSE: the official result names the YES side (Tegmark and Bengio) the winner or calls it a tie.

Evidence

  1. Artificial Intelligence Debate Munk Debates · 22 June 2023
    “Swing: -3% (Con team win)”
  2. The Hub: AI poses an existential threat, according to Munk Debates crowd, although a few were persuaded otherwise - Munk Debates Munk Debates · 28 June 2023
    “only 3 percent of the Munk Debate audience changed their opinions by the end of the debate.”
  3. Did Bengio and Tegmark lose a debate about AI x-risk against LeCun and Mitchell? — LessWrong LessWrong · 25 June 2023
    “Con wins by a 4% gain (probably not 3% due to rounding)”
  4. Did Bengio and Tegmark lose a debate about AI x-risk against LeCun and Mitchell? — LessWrong LessWrong · 25 June 2023
    “Pre-debate: 67% Pro, 33% Con”

Reasoning

The Munk Debates' official page lists the result as "Swing: -3% (Con team win)" (E1), so the NO side (Mitchell and LeCun) is the official winner on the vote-swing measure. The official Munk Debates page gives the swing as 3 points, not the 4 the claim states (E1). A LessWrong comment (E3) reports a 4% gain for the Con side and says it is probably not 3% because of rounding, but it is not the official source. Because the margin is off by about one point, the claim is mostly true rather than true.

Cite this claim

Yann LeCun, “In the official result of the Munk Debate on AI existential threat, the NO side (Mitchell and LeCun) won by 4% over the YES side (Tegmark and Bengio).” (said 28 June 2023). Verdict: mostly true. Reliability API, claim c_d441ba74750a, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_d441ba74750a

Verdicts are made automatically; how verdicts and scores are made. The same data is on Hugging Face under CC BY 4.0.