ClaimsPeople

Nate Silver

Nate Silver is an election forecaster and writer who founded FiveThirtyEight and now publishes the Silver Bulletin newsletter.

617 claims collected, 163 researched, 84 resolved. Report of 10 November 2026, method 0.1.0+28da144b.

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?

    82 76 88

    High confidence, 63 factual claims resolved. From 19 documents, 2012 to 2026; 19 documents sampled from 35 candidates found.

    Not scored: 42 unresolvable, 7 checkable only against their own records.

  • Predictions

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

    61 47 75

    Medium confidence, 16 predictions resolved. From 8 documents, 2015 to 2025; 19 documents sampled from 35 candidates found.

    Not scored: 12 unresolvable. 10 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?

    67 46 85

    Low confidence, 5 promises resolved. From 4 documents, 2022 to 2025; 19 documents sampled from 35 candidates found.

    Not scored: 7 unresolvable.

All types combined: 77 (likely 72 to 82, high confidence), from 84 claims: 63 factual claims, 16 predictions, 5 promises. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.

Nate Silver's claims

37 claims

37 claims, sorted by newest said. Choose a claim to read it in full.
VerdictKindSaidClaimTopic
Mostly true Factual 28 Oct 2026 Abdul El-Sayed is not a DSA guy but is sort of a more left populist. Factual · 28 Oct 2026 AI policy and governance
Mostly true Factual 23 Oct 2026 Jocelyn Benson's lead in Michigan gubernatorial polls is about 4.2 points larger than Abdul El-Sayed's lead in the Michigan Senate polls. Factual · 23 Oct 2026 AI policy and governance
Mostly true Factual 8 Apr 2025 The average US tariff rate on the rest of the world was at least 23% as of April 2025. Factual · 8 Apr 2025 Economics and markets
Mostly true Factual 8 Apr 2025 The two special elections held in early April 2025 were in extremely Republican-leaning, rural Florida districts. Factual · 8 Apr 2025 AI policy and governance
Mostly true 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. Factual · 7 Nov 2024 Other
Mostly true Factual 9 Oct 2024 Silver said that, if anything, there was too little panic among Democrats about 2016, and that they were dismissive of polls showing a tightening race late in the campaign and a tenuous Hillary Clinton lead in the Electoral College. Factual · 9 Oct 2024 Philosophy, ethics and society
Mostly true Factual 9 Oct 2024 The previous Quinnipiac polls of Michigan, Wisconsin and Pennsylvania were quite a bit better for Harris than the latest ones. Factual · 9 Oct 2024 Science and research methods
Mostly true Factual 9 Oct 2024 Since Nate Silver began covering elections in 2008, Democrats have been prone to bouts of panic about the state of the horse race. Factual · 9 Oct 2024 Philosophy, ethics and society
Mostly true Factual 8 Oct 2024 In baseball, single-run leads are relatively durable because runs tend to come in bunches and the default outcome is not to score at all. Factual · 8 Oct 2024 Science and research methods
Mostly true Factual 13 Aug 2024 Humans evolved over hundreds of thousands of years in an environment without abundance. Factual · 13 Aug 2024 Philosophy, ethics and society
Mostly true Factual 13 Aug 2024 In August 2024, Silver suggested, as a rhetorical question, that AI still could not fold clothes or do laundry. Factual · 13 Aug 2024 AI capabilities
Mostly true Prediction 13 Aug 2024 Kamala Harris's bid is inherently difficult and a little bit of a long shot, but her chances of winning the 2024 presidential election are probably better than Joe Biden's were. Prediction · 13 Aug 2024 AI timelines and forecasting
Mostly true Promise 11 Sep 2023 Silver Bulletin will publish occasional paywalled posts on other topics for paid subscribers. Promise · 11 Sep 2023 Their own work or organisation
Mostly true Factual 11 Sep 2023 Nate Silver said launching FiveThirtyEight was basically his fourth career, after management consulting, professional poker, and baseball analytics. Factual · 11 Sep 2023 Their own work or organisation
Mostly true Factual 11 Sep 2023 On Substack, subscriptions automatically renew at the price the subscriber signed up for. Factual · 11 Sep 2023 Economics and markets
Mostly true Factual 8 Sep 2023 The Wall Street Journal polls in partnership between a Republican polling firm and a Democratic polling firm. Factual · 8 Sep 2023 Science and research methods
Mostly true Factual 21 May 2022 Joe Biden's approval rating never had a spike. Factual · 21 May 2022 AI timelines and forecasting
Mostly true Prediction 29 Jun 2021 It seems very likely that people who received one dose of the J&J COVID vaccine would benefit from a Pfizer or Moderna booster. Prediction · 29 Jun 2021 Biosecurity
Mostly true Factual 16 Jun 2021 The risk of a vaccinated young person getting severely ill from COVID is incredibly low, though not zero. Factual · 16 Jun 2021 Biosecurity
Mostly true Factual 8 Jun 2021 The 2020 election forecasts by Nate Silver's team were largely accurate and well-calibrated despite a bad year for the polls. Factual · 8 Jun 2021 Their own work or organisation
Mostly true Prediction 28 Mar 2020 The coronavirus peak in the NYC metro area will probably come sooner than elsewhere in the US. Prediction · 28 Mar 2020 Biosecurity
Mostly true Factual 29 Sep 2019 Margins of error in individual state and national primary polls on support among black or Hispanic voters are often very high, for example around +/- 10%. Factual · 29 Sep 2019 Science and research methods
Mostly true Factual 26 Sep 2019 In the Democratic primary, a candidate who gets under 15% in a given state or district receives no delegates. Factual · 26 Sep 2019 AI policy and governance
Mostly true Prediction 25 Sep 2019 Nate Silver guessed that the public would see the contents of the Ukraine call transcript as very inappropriate conduct and would find the White House's quid pro quo spin unresponsive to their concerns. Prediction · 25 Sep 2019 AI timelines and forecasting
Mostly true Prediction 23 Sep 2019 Nate Silver said he still tended not to think public opinion on impeaching Trump would move much, but that it was hard to know. Prediction · 23 Sep 2019 AI timelines and forecasting
Mostly true Factual Nov 2016 The main reason FiveThirtyEight's 2016 model gave Trump a better chance than other models was its assumption that polling errors are correlated. Factual · Nov 2016 AI timelines and forecasting
Mostly true Factual Nov 2016 Clinton significantly underperformed her polls across the Midwest and Rust Belt in the 2016 election, costing her the Electoral College. Factual · Nov 2016 AI timelines and forecasting
Mostly true Factual 11 Oct 2016 The two scientific polls after the second 2016 presidential debate showed a Clinton win, but a more marginal one than after the first debate. Factual · 11 Oct 2016 Science and research methods
Mostly true Factual 11 Oct 2016 Silver said that all the scientific polls after the first 2016 presidential debate showed Clinton had won by a pretty emphatic margin. Factual · 11 Oct 2016 Science and research methods
Mostly true Factual 2016 The t-distribution in FiveThirtyEight's model mostly makes a difference for very low-probability events. Factual · 2016 Science and research methods
Mostly true Factual 2016 In 2012, Obama beat his polling by 2 or 3 percentage points in almost every swing state. Factual · 2016 Science and research methods
Mostly true Factual 2016 FiveThirtyEight's model is based on a combination of state and national polls. Factual · 2016 Science and research methods
Mostly true Factual 2016 FiveThirtyEight's polls-only model gives Clinton an 85 percent chance of winning and its polls-plus model gives her 83 percent. Factual · 2016 Science and research methods
Mostly true Factual 2015 Donald Trump's overall favorability ratings in 2015 are about 30 percent favorable and 60 percent unfavorable, and have not improved. Factual · 2015 Other
Mostly true Factual 2015 In presidential primaries, endorsements and party elite support tend to be more reliable indicators of eventual success than early polls. Factual · 2015 Other
Mostly true Factual 2015 Of Lieberman, Clinton, Giuliani and Perry, only Clinton came close to winning the nomination. Factual · 2015 Other
Mostly true Factual 2015 Even this far out, national primary polls have some correlation with outcomes: a candidate near the top is a somewhat better bet to win the nomination than one near the bottom. Factual · 2015 Science and research methods
Back to the list

Nate Silver

Open its page

Factual AI capabilities

“It still can't fold clothes, right?”

Nate Silver, interviewee 13 August 2024 Nate Silver's On the Edge audio interview (Audible)

In August 2024, Silver suggested, as a rhetorical question, that AI still could not fold clothes or do laundry.

Mostly true True in substance, wrong in a detail.

What would count as true or false

Judged as of 2024-08-13. The claim is that AI systems (AI-driven robots) still could not fold clothes. True: evidence from mid-2024 shows that no AI or robotic system could fold ordinary clothes in a general, practical way. Isolated research demos of narrow folding, on specific items in controlled settings, do not change this, because the remark is about general capability. Mostly true: general capability was still absent, but credible research demos of AI-driven robots folding clothes (e.g. towels or shirts) did exist by August 2024, which the remark glosses over. Mixed: experts or the evidence from that time are split on whether AI robots could fold clothes in a meaningful sense. Mostly false: by August 2024, AI-driven robots had publicly shown reliable folding of varied real garments, beyond narrow staged demos, even though they were not commercially deployed. False: by August 2024, AI-powered systems were deployed or sold that reliably fold everyday clothes.

Evidence

  1. Learning Arbitrary-Goal Fabric Folding with One Hour of Real Robot Experience PMLR (Proceedings of Machine Learning Research), Conference on Robot Learning · 4 October 2021
    “We demonstrate our method on a set of towel-folding tasks, and show that our approach is able to discover sequential folding strategies, purely from trial-and-error.”
  2. Learning Keypoints from Synthetic Data for Robotic Cloth Folding arXiv
    “We evaluate the performance of this detector for folding towels on a unimanual robot setup and find that the grasp and fold success rates are 77% and 53%, respectively.”
  3. Learning Keypoints from Synthetic Data for Robotic Cloth Folding arXiv
    “The fold success rate is lower than the grasp success rate due to the limitations of scripted trajectories and single-arm manipulation:”
  4. Learning Keypoints from Synthetic Data for Robotic Cloth Folding arXiv
    “However, more extensive tuning is required to completely overcome the reality gap.”
  5. Learning Keypoints from Synthetic Data for Robotic Cloth Folding arXiv (academic preprint by university robotics researchers)
    “We evaluate the performance of this detector for folding towels on a unimanual robot setup and find that the grasp and fold success rates are 77% and 53%, respectively.”
  6. Learning Keypoints from Synthetic Data for Robotic Cloth Folding arXiv (academic preprint by university robotics researchers)
    “Although these systems perform very well in their test settings, the segmentation algorithms assume a known background color distribution, which strongly limits the potential to generalise to diverse environments.”
  7. Learning Keypoints from Synthetic Data for Robotic Cloth Folding arXiv (academic preprint by university robotics researchers)
    “Although significant progress has been made, cloth manipulation remains a challenging task in terms of perception and control”
  8. SpeedFolding: Learning Efficient Bimanual Folding of Garments alphaXiv (arXiv paper 2208.10552, Avigal et al.) · 9 September 2022
    “After learning from 4300 human-annotated and self-supervised actions, the robot is able to fold garments from a random initial configuration in under 120s on average with a success rate of 93%.”
  9. FabricFolding: learning efficient fabric folding without expert demonstrations | Robotica | Cambridge Core Cambridge University Press (Robotica journal) · 4 April 2024
    “Specifically, our method achieves a coverage rate of 0.822 and a success rate of 0.88 for long-sleeved T-shirts folding.”
  10. Folding laundry: Humanoid Robot Evidence humanoidversus · 19 March 2024
    “The EVE update includes shirt folding. The maker's video description says all actions are autonomous and shown at normal speed; no fold-quality or repeatability metric is given.”
  11. Folding laundry: Humanoid Robot Evidence humanoidversus · 15 January 2024
    “Electrek quotes Tesla engineer Milan Kovac describing teleoperation for dexterity verification and training-data collection.”

Reasoning

Before August 2024, research papers showed robots folding towels and garments in lab settings, including 93% success on garments in SpeedFolding (E8) and 88% on long-sleeved T-shirts in FabricFolding (E9), plus earlier towel-folding work (E1, E2). A humanoid maker's March 2024 update showed autonomous shirt folding but gave no figures for quality or repeatability (E10). Researchers described cloth manipulation as still challenging and said generalisation was limited (E6, E7). None of the evidence shows deployed or commercial systems that reliably fold everyday clothes, so general capability was still absent, but the remark glosses over credible research demos.

Cite this claim

Nate Silver, “In August 2024, Silver suggested, as a rhetorical question, that AI still could not fold clothes or do laundry.” (said 13 August 2024). Verdict: mostly true. Reliability API, claim c_acab987c84e9, method 0.1.0+28da144b. https://epistacklabs.com/claims/c_acab987c84e9

Claims without a verdict

Is a verdict here wrong, or do you speak for Nate Silver? 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.