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
Not checkable Dario Amodei Not checkable 11 Nov 2024 I have an instinct that there’s a lot of room at the top for AIs to get smarter. Dario Amodei · Not checkable · 11 Nov 2024
Not checkable Dario Amodei Not checkable 11 Nov 2024 the biggest problem of biology is that you can’t see what’s going on. Dario Amodei · Not checkable · 11 Nov 2024
Not researched Dario Amodei Factual 11 Nov 2024 Anthropic showed demos of Claude filling out spreadsheets, interacting with websites and opening programs on Windows, Linux and Mac. Dario Amodei · Factual · 11 Nov 2024 AI capabilities
Not checkable Dario Amodei Not checkable 11 Nov 2024 Moore’s laws, scaling laws, they’re not laws of the universe. Dario Amodei · Not checkable · 11 Nov 2024
Not yet due Dario Amodei Prediction 11 Nov 2024 Within two to three years, compute clusters will be large enough to deploy millions of instances of AI models. Dario Amodei · Prediction · 11 Nov 2024 AI compute, hardware and data
Unresolvable Dario Amodei Promise 11 Nov 2024 Anthropic will release models more powerful than the models that exist today. Dario Amodei · Promise · 11 Nov 2024 Their own work or organisation
Not checkable Dario Amodei Not checkable 11 Nov 2024 often the level of intelligence may just not be the limiting factor. Dario Amodei · Not checkable · 11 Nov 2024
Unresolvable Dario Amodei Prediction 11 Nov 2024 AI scaling will probably continue to produce gains. Dario Amodei · Prediction · 11 Nov 2024 AI timelines and forecasting
Not checkable Dario Amodei Not checkable 11 Nov 2024 I think humanity has been protected by the fact that the overlap between really smart, well-educated people and people who want to do… Dario Amodei · Not checkable · 11 Nov 2024
Not checkable Dario Amodei Not checkable 11 Nov 2024 I can’t speak to whether we have better tooling than other companies. Dario Amodei · Not checkable · 11 Nov 2024
Not yet due Dario Amodei Prediction 11 Nov 2024 Within two to three years, the millions of deployable AI model instances will run faster than humans. Dario Amodei · Prediction · 11 Nov 2024 AI capabilities
Not checkable Dario Amodei Not checkable 11 Nov 2024 The system prompt can have some effects, although it’s unlikely to dumb down models, it’s unlikely to make them dumber. Dario Amodei · Not checkable · 11 Nov 2024
Own records only Dario Amodei Factual 11 Nov 2024 In 2017, seeing GPT-1 results convinced Amodei that language was probably the area where scaling could work. Dario Amodei · Factual · 11 Nov 2024 Their own work or organisation
Not checkable Dario Amodei Not checkable 11 Nov 2024 It’s all a set of imperfect people trying to aim imperfectly at some ideal that will never perfectly be achieved. Dario Amodei · Not checkable · 11 Nov 2024
Not checkable Dario Amodei Not checkable 11 Nov 2024 It is very difficult to control across the board how the models behave. Dario Amodei · Not checkable · 11 Nov 2024
Not checkable Dario Amodei Not checkable 11 Nov 2024 my worry is that by being a much more intelligent agent, AI could break that correlation. Dario Amodei · Not checkable · 11 Nov 2024
Not checkable Dario Amodei Not checkable 11 Nov 2024 It’s only possible to do things so fast in the physical world. Dario Amodei · Not checkable · 11 Nov 2024
Not researched Dario Amodei Factual 11 Nov 2024 On typical real-world programming tasks, AI models went from 3% in January 2024 to 50% in October 2024. Dario Amodei · Factual · 11 Nov 2024 AI capabilities
Not checkable Dario Amodei Not checkable 11 Nov 2024 It’s something we struggle with surprisingly much relative to how trivial it is for the grand science of training the models. Dario Amodei · Not checkable · 11 Nov 2024
Not checkable Dario Amodei Not checkable 11 Nov 2024 Not giving you an exact date, but as far as we know, the plan is still to have a Claude 3.5 Opus. Dario Amodei · Not checkable · 11 Nov 2024
Not checkable Dario Amodei Not checkable 11 Nov 2024 We can’t have a small group of people who are developing these systems say, “This is what’s best for everyone.” Dario Amodei · Not checkable · 11 Nov 2024
Not checkable Dario Amodei Not checkable 11 Nov 2024 It’s very easy to go from a hundred to a thousand, a thousand to 10,000 without paying attention to making sure everyone has a unified… Dario Amodei · Not checkable · 11 Nov 2024
Not checkable Dario Amodei Not checkable 11 Nov 2024 the whole history of biology, a whole big part of the history is basically our ability to read and understand what’s going on and our… Dario Amodei · Not checkable · 11 Nov 2024
Not checkable Dario Amodei Not checkable 11 Nov 2024 I worry about economics and the concentration of power. That’s actually what I worry about more. Dario Amodei · Not checkable · 11 Nov 2024
Not checkable Dario Amodei Not checkable 11 Nov 2024 I think we haven't done perfectly on them. I think we've actually done the best of all the AI companies, but still so far from perfect. Dario Amodei · Not checkable · 11 Nov 2024
Resolved true Dario Amodei Factual 11 Nov 2024 DeepMind's AlphaGo Zero required no example data from humans. Dario Amodei · Factual · 11 Nov 2024 AI capabilities
Not checkable Dario Amodei Not checkable 11 Nov 2024 We're getting better and better at solving this. I don't think this is an unsolvable problem. Dario Amodei · Not checkable · 11 Nov 2024
Not researched Gary Marcus Factual 9 Nov 2024 Gary Marcus has warned about foundational limits of neural networks since his 2001 book The Algebraic Mind, where he first described hallucinations. Gary Marcus · Factual · 9 Nov 2024 Their own work or organisation
Not researched Gary Marcus Factual 9 Nov 2024 Leading AI developers' LLMs are converging to roughly the same level of capability, leaving no company with a competitive moat. Gary Marcus · Factual · 9 Nov 2024 Economics and markets
Not checkable Gary Marcus Not checkable 9 Nov 2024 As I have always warned, that’s just a fantasy. Gary Marcus · Not checkable · 9 Nov 2024
Unresolvable Gary Marcus Prediction 9 Nov 2024 The US's focus on LLMs for AI may prove to be a massive mistake. Gary Marcus · Prediction · 9 Nov 2024 AI policy and governance
Unresolvable Gary Marcus Prediction 9 Nov 2024 LLMs will become a commodity and price wars will keep revenue low. Gary Marcus · Prediction · 9 Nov 2024 Economics and markets
Not researched Sam Altman Factual 9 Nov 2024 According to Similarweb, ChatGPT is the 8th biggest website in the world as of November 2024. Sam Altman · Factual · 9 Nov 2024 Technology
Not researched Gary Marcus Factual 9 Nov 2024 Scaling of AI models is running out of steam, with improvements from adding data and compute diminishing. Gary Marcus · Factual · 9 Nov 2024 AI timelines and forecasting
Not checkable Sam Altman Not checkable 9 Nov 2024 we are much, much better at reliability than we used to be Sam Altman · Not checkable · 9 Nov 2024
Unresolvable Gary Marcus Prediction 9 Nov 2024 LLMs will not disappear, but their economics will likely never make sense because training costs rise with scaling. Gary Marcus · Prediction · 9 Nov 2024 AI, the economy and work
Not checkable Gary Marcus Not checkable 9 Nov 2024 There is no principled solution to hallucinations in systems that traffic only in the statistics of language without explicit… Gary Marcus · Not checkable · 9 Nov 2024
Unresolvable Gary Marcus Prediction 9 Nov 2024 Given chip costs, profits from LLMs will be elusive. Gary Marcus · Prediction · 9 Nov 2024 Economics and markets
Not checkable Sam Altman Not checkable 9 Nov 2024 clearly more work in front of us. Sam Altman · Not checkable · 9 Nov 2024
Unresolvable Gary Marcus Prediction 9 Nov 2024 The AI financial bubble may burst quickly once people realize the situation, and Nvidia's valuation may take a hit. Gary Marcus · Prediction · 9 Nov 2024 Economics and markets
Not checkable Gary Marcus Not checkable 9 Nov 2024 I’m glad that the market is finally recognizing that what I’ve been saying is true. Gary Marcus · Not checkable · 9 Nov 2024
Unresolvable Gary Marcus Factual 9 Nov 2024 Melanie Mitchell, Subbarao Kambhampati, Emily Bender, Ernie Davis and others have long emphasized that LLMs have principled limits. Gary Marcus · Factual · 9 Nov 2024 Their own work or organisation
Not checkable Gary Marcus Not checkable 9 Nov 2024 The thing is, in the long term, science isn’t majority rule. In the end, the truth generally outs. Gary Marcus · Not checkable · 9 Nov 2024
Resolved true Gary Marcus 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. Gary Marcus · Factual · 9 Nov 2024 Their own work or organisation
Not researched Nate Silver Factual 7 Nov 2024 Trump won more than 312 electoral votes in 16 percent of the model's final simulations. Nate Silver · Factual · 7 Nov 2024 Their own work or organisation
Unresolvable Nate Silver Factual 7 Nov 2024 In Silver Bulletin's final model, Harris won all swing states except Arizona in about 4 percent of simulations. Nate Silver · Factual · 7 Nov 2024 Their own work or organisation
Unresolvable Nate Silver Prediction 7 Nov 2024 Noting that many mail votes were still to be counted, Silver guessed that most of the shift toward Democrats in Washington, Oregon and Colorado would hold up. Nate Silver · Prediction · 7 Nov 2024 Other
Not checkable Nate Silver Not checkable 7 Nov 2024 I’m not sure exactly what outcome I was rooting for on Tuesday night. Nate Silver · Not checkable · 7 Nov 2024
Mixed Nate Silver Factual 7 Nov 2024 Utah has gradually become more politically purple, due to a tech-sector workforce and Mormons not always being keen on Trump. Nate Silver · Factual · 7 Nov 2024 Other
Not researched Nate Silver Factual 7 Nov 2024 Silver said Harris lost Georgia by 2 points, and that this means Georgia is now closer to the tipping point. Nate Silver · Factual · 7 Nov 2024 Other
Not researched Nate Silver Factual 7 Nov 2024 Trump winning exactly 312 electoral votes was by far the most common outcome in the model's simulations, occurring 6 percent of the time. Nate Silver · Factual · 7 Nov 2024 Their own work or organisation
Not checkable Nate Silver Not checkable 7 Nov 2024 Still, it’s a sign that the model’s internal logic is sound, and that American elections are quite predictable on a state level once you… Nate Silver · Not checkable · 7 Nov 2024
Unresolvable Nate Silver Prediction 7 Nov 2024 The model's other 48 of 50 state calls will be right, assuming Trump's leads in Arizona and Nevada hold, which Silver said they almost certainly will. Nate Silver · Prediction · 7 Nov 2024 Their own work or organisation
Not researched Nate Silver Factual 7 Nov 2024 In Silver Bulletin's model, states with many polls are weighted almost entirely on polling averages, while states without polls default to 2020 results plus a uniform shift toward Trump. Nate Silver · Factual · 7 Nov 2024 Their own work or organisation
Not checkable Nate Silver Not checkable 7 Nov 2024 We’re treating next week (and the remainder of this week) as “election wind-down mode.” Nate Silver · Not checkable · 7 Nov 2024
Not researched Nate Silver Factual 7 Nov 2024 The actual 2024 election map was the most common map in Silver Bulletin's final model, appearing in 4,660 of 80,000 simulations. Nate Silver · Factual · 7 Nov 2024 Their own work or organisation
Not checkable Nate Silver Not checkable 7 Nov 2024 Silver Bulletin, we hope, will have a long future. Nate Silver · Not checkable · 7 Nov 2024
Not checkable Nate Silver Not checkable 7 Nov 2024 So no matter what, actually, we were going to be wrong about something! Nate Silver · Not checkable · 7 Nov 2024
Not researched Nate Silver Factual 7 Nov 2024 Final 2024 presidential margins in the swing states were close to the polling averages, and less close in blue states. Nate Silver · Factual · 7 Nov 2024 Science and research methods
Not checkable Nate Silver Not checkable 7 Nov 2024 Democrats, however — and here, I’m not referring so much to Silver Bulletin subscribers but in the broader universe online — often get… Nate Silver · Not checkable · 7 Nov 2024
Not yet due Nate Silver Prediction 7 Nov 2024 Utah could become a swing state by 2040. Nate Silver · Prediction · 7 Nov 2024 Other
Not checkable Nate Silver Not checkable 7 Nov 2024 I support third-party ballot access, and stronger performance helps third parties to qualify for the New York ballot in future elections. Nate Silver · Not checkable · 7 Nov 2024
Unresolvable Nate Silver Promise 7 Nov 2024 Silver Bulletin will publish about 9-12 posts a month after the election wind-down. Nate Silver · Promise · 7 Nov 2024 Their own work or organisation
Not checkable Nate Silver Not checkable 7 Nov 2024 although honestly, I think I’d still have taken some crap from Democrats for any Harris win. Nate Silver · Not checkable · 7 Nov 2024
Not checkable Nate Silver Not checkable 7 Nov 2024 When polls miss low on Trump in one key state, they probably also will in most or all of the others. Nate Silver · Not checkable · 7 Nov 2024
Not checkable Nate Silver Not checkable 7 Nov 2024 Starting the week of Nov. 17, this newsletter will be more like it was before the election, with a greater breadth of topics. Nate Silver · Not checkable · 7 Nov 2024
Mostly false Nate Silver Promise 7 Nov 2024 Silver said he would probably begin with college basketball Elo ratings, with a goal of launching by the end of 2024, if not sooner. Nate Silver · Promise · 7 Nov 2024 Their own work or organisation
Unresolvable Nate Silver Promise 7 Nov 2024 Silver Bulletin will make all its posts free during the post-election wind-down period, except a bonus subscriber questions post. Nate Silver · Promise · 7 Nov 2024 Their own work or organisation
Unresolvable Nate Silver Prediction 7 Nov 2024 Voters leaving deep blue states while Republicans gain surplus voters in Florida and Texas will greatly reduce and perhaps reverse the Electoral College bias favoring Republicans. Nate Silver · Prediction · 7 Nov 2024 Other
Not checkable Nate Silver Not checkable 7 Nov 2024 That’s not a good way to build a majority, and now Democrats no longer have one. Nate Silver · Not checkable · 7 Nov 2024
Not checkable Nate Silver Not checkable 7 Nov 2024 Republicans are generally happy when you agree with them partway or half the time. Nate Silver · Not checkable · 7 Nov 2024
Not checkable Nate Silver Not checkable 7 Nov 2024 Trump was happy to take on all comers, whereas with Democrats, disagreement on any hot-button topic (say, COVID school closures or Biden’s… Nate Silver · Not checkable · 7 Nov 2024
Not researched Nate Silver Factual 7 Nov 2024 In Silver Bulletin's final model, a Harris sweep of the swing states was the second most common outcome, occurring 14 percent of the time. Nate Silver · Factual · 7 Nov 2024 Their own work or organisation
Own records only Nate Silver Factual 7 Nov 2024 Nate Silver voted for Kamala Harris in the 2024 presidential election. Nate Silver · Factual · 7 Nov 2024 Their own work or organisation
Not researched Nate Silver Factual 7 Nov 2024 Silver Bulletin's model, read deterministically, had Trump winning 5 of 7 swing states and missed only Michigan and Wisconsin, where Trump leads by 1.5 and 0.9 points in current counts. Nate Silver · Factual · 7 Nov 2024 Their own work or organisation
Mostly true Nate Silver Factual 7 Nov 2024 Conservatives leaving blue states tend to move to Florida and Texas, making those states redder, and many polls did not catch up. Nate Silver · Factual · 7 Nov 2024 Other
Not checkable Nate Silver Not checkable 7 Nov 2024 we plan to reinvest in Silver Bulletin for the long term. Nate Silver · Not checkable · 7 Nov 2024
Unresolvable Nate Silver Promise 7 Nov 2024 Silver said the plan was for Silver Bulletin to have one paid post and one free post a week, in addition to SBSQ subscriber-questions posts. Nate Silver · Promise · 7 Nov 2024 Their own work or organisation
Unresolvable Nate Silver Factual 7 Nov 2024 In simulations producing the actual map, projected state margins missed actual results by an average of 2.3 points, given results reported so far. Nate Silver · Factual · 7 Nov 2024 Their own work or organisation
Not checkable Nate Silver Not checkable 7 Nov 2024 Tuesday was a perfect demonstration of correlated polling error Nate Silver · Not checkable · 7 Nov 2024
Not checkable Nate Silver Not checkable 7 Nov 2024 So I essentially got to perform a randomized control trial on how partisans in both camps reacted to good and bad news. Nate Silver · Not checkable · 7 Nov 2024
Not researched Nate Silver Factual 7 Nov 2024 New Mexico remained solidly Democratic in the 2024 presidential election. Nate Silver · Factual · 7 Nov 2024 Other
Unresolvable Nate Silver Factual 7 Nov 2024 Nate Silver said Iowa stood apart as its own group because of the Selzer poll, which he said was not highly accurate. Nate Silver · Factual · 7 Nov 2024 Science and research methods
Resolved true Nate Silver Factual 7 Nov 2024 In Silver Bulletin's final model, a Trump sweep of the seven swing states was the most common scenario, occurring in 20 percent of simulations. Nate Silver · Factual · 7 Nov 2024 Their own work or organisation
Not checkable Jeff Bezos Not checkable 28 Oct 2024 What presidential endorsements actually do is create a perception of bias. A perception of non-independence. Jeff Bezos · Not checkable · 28 Oct 2024
Not checkable Jeff Bezos Not checkable 28 Oct 2024 The stakes are too high. Jeff Bezos · Not checkable · 28 Oct 2024
Own records only Jeff Bezos Factual 28 Oct 2024 No quid pro quo of any kind was involved in the Washington Post's decision not to endorse a presidential candidate in 2024. Jeff Bezos · Factual · 28 Oct 2024 Their own work or organisation
Not checkable Jeff Bezos Not checkable 28 Oct 2024 No undecided voters in Pennsylvania are going to say, ‘I’m going with Newspaper A’s endorsement.’ None. Jeff Bezos · Not checkable · 28 Oct 2024
Not checkable Jeff Bezos Not checkable 28 Oct 2024 Ending them is a principled decision, and it’s the right one. Jeff Bezos · Not checkable · 28 Oct 2024
Not checkable Jeff Bezos Not checkable 28 Oct 2024 because I knew it would provide ammunition to those who would like to frame this as anything other than a principled decision. Jeff Bezos · Not checkable · 28 Oct 2024
Not checkable Jeff Bezos Not checkable 28 Oct 2024 complexifier for the Post. Jeff Bezos · Not checkable · 28 Oct 2024
Not checkable Jeff Bezos Not checkable 28 Oct 2024 I wish we had made the change earlier than we did, in a moment further from the election and the emotions around it, Jeff Bezos · Not checkable · 28 Oct 2024
Unresolvable Jeff Bezos Factual 28 Oct 2024 Jeff Bezos says the Washington Post's late decision to stop endorsing presidential candidates resulted from inadequate planning, not an intentional strategy. Jeff Bezos · Factual · 28 Oct 2024 Their own work or organisation
Not checkable Jeff Bezos Not checkable 28 Oct 2024 a bulwark against intimidation Jeff Bezos · Not checkable · 28 Oct 2024
Unresolvable Jeff Bezos Factual 28 Oct 2024 There was no connection between the Trump meeting with Blue Origin executives and the Washington Post's decision not to endorse. Jeff Bezos · Factual · 28 Oct 2024 Their own work or organisation
Unresolvable Jeff Bezos Factual 28 Oct 2024 Jeff Bezos did not know beforehand about the Trump meeting with Blue Origin executives held on the day the Post announced its non-endorsement. Jeff Bezos · Factual · 28 Oct 2024 Their own work or organisation
Not checkable Jeff Bezos Not checkable 28 Oct 2024 appearance of conflict Jeff Bezos · Not checkable · 28 Oct 2024
Own records only Jeff Bezos Factual 28 Oct 2024 Neither presidential campaign nor candidate was consulted or informed about the Washington Post's non-endorsement decision, which was made entirely internally. Jeff Bezos · Factual · 28 Oct 2024 Their own work or organisation
Not checkable Jeff Bezos Not checkable 28 Oct 2024 Presidential endorsements do nothing to tip the scales of an election, Jeff Bezos · Not checkable · 28 Oct 2024
Not checkable Jeff Bezos Not checkable 28 Oct 2024 While I do not and will not push my personal interest, I will also not allow this paper to stay on autopilot and fade into irrelevance —… Jeff Bezos · Not checkable · 28 Oct 2024
Back to the list

Nate Silver

Open its page

Not checkable

“When polls miss low on Trump in one key state, they probably also will in most or all of the others.”

Nate Silver 7 November 2024 The model exactly predicted the most likely election map (Silver Bulletin)

When polls miss low on Trump in one key state, they probably also will in most or all of the others.

Not checkable An opinion, value judgement or statement with nothing to check.

Cite this claim

Nate Silver, “When polls miss low on Trump in one key state, they probably also will in most or all of the others.” (said 7 November 2024). Verdict: not checkable. Reliability API, claim c_5081482dda1e, method 0.1.0+28da144b. https://epistacklabs.com/claims/c_5081482dda1e

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