Start with one
Michael Burry predicted, 5 April 2011:
Burry said he expected the US government to attempt to continue easy money policies into the next presidential term.
“Today, I expect the US government to attempt to continue easy money policies into the next presidential term”
Did it come true? Show the verdict
Resolved true The record shows it happened or was so.
How to read a verdict
Each claim is one thing a person said in public, quoted word for word. The verdict says how it held up against the public record.
- Resolved true The record shows it happened or was so.
- Mostly true True in substance, wrong in a detail.
- Mixed Partly true and partly false in roughly equal measure.
- Mostly false False in substance, right in a detail.
- Resolved false The record shows it did not happen or was not so.
Every checked claim, person by person
Each bar is one person's checked claims, coloured by verdict. Choose a name to read them.
- Balaji Srinivasan 80
- Bill Ackman 99
- Bill Gates 85
- Daniela Amodei 88
- Dario Amodei 97
- Demis Hassabis 86
- Eliezer Yudkowsky 79
- Elon Musk 94
- Gary Marcus 103
- GiveWell 23
- Jeff Bezos 96
- Jensen Huang 107
- Jeremy Grantham 73
- Jim Cramer 14
- Marc Andreessen 105
- Mark Zuckerberg 90
- Michael Burry 73
- Nassim Taleb 94
- Nate Silver 84
- Nouriel Roubini 66
- Ørsted 29
- Peter Thiel 55
- Ray Dalio 91
- Ray Kurzweil 78
- Rodney Brooks 88
- Sam Altman 87
- Satya Nadella 98
- Scott Alexander 81
- Steven Pinker 99
- Tyler Cowen 97
- Vinod Khosla 52
- Warren Buffett 78
- Yann LeCun 84
- Resolved true 973
- Mostly true 1,191
- Mixed 244
- Mostly false 183
- Resolved false 62
Looking for something?
What the record shows
Each finding is computed by a fixed rule from the published claims and appears only when its rule holds.
40 of the 62 claims resolved false are predictions, though predictions are 340 of the 2,653 resolved claims.
Of 340 resolved predictions, 227 came true or mostly true, 81 resolved false or mostly false, and 32 were mixed.
The largest gaps between a person's own factual and prediction scores, where both are at medium or high confidence: Nouriel Roubini: factual 72 (48 resolved), predictions 40 (18 resolved, medium confidence); Elon Musk: factual 70 (35 resolved), predictions 46 (21 resolved, medium confidence); Nate Silver: factual 82 (63 resolved), predictions 61 (16 resolved, medium confidence).
Most resolved claims by topic: Their own work or organisation (666), Economics and markets (567), AI capabilities (256).
73 resolved claims are tagged political and are scored like any other; the most come from Nate Silver (27).
The two model judges differed on 1,249 of 2,653 resolved claims: in 962 the second judge gave no verdict, and in 20 the two were two or more steps apart. Agreement between judges is consistency, not accuracy.
Verdicts are made automatically and are not individually reviewed by a person. How claims are checked. Live figures: they follow corrections made since the last dataset release. Reading is open: the dataset, the claims and single lookups need no key and no sign-up. Running reports on new subjects, the full API and the bulk export are by invitation, for design partners. For subjects: how to reply. Download the data.