16h ago

Allan Dafoe says the $20,000 Kurzweil-Kapor Turing Test bet protocol has a 44% false negative rate

Actual humans fail the test nearly half the time

——0——
Original post
OPMatthew Barnett#1343Matthew Barnett@MATTHEWJBAR

You can read the detailed terms here: https://longbets.org/1/#adjudication_terms

12:56 PM · May 23, 2026 View on X
REPLYAllan Dafoe#1127Allan Dafoe@ALLANDAFOE

@MatthewJBar Interesting specificity. Worth noting that the test has a ~44% false negative rate. A real human, indistinguishable from the others, would fail it almost half the time.

Matthew BarnettMatthew Barnett@MatthewJBar

In 2002, Ray Kurzweil and Mitchell Kapor made a bet on whether the Turing Test would be passed by 2029. To determine the winner, they devised detailed terms, including the appointment of a committee to administer an actual exam, with a set of judges and human foils.

7:55 PM · May 23, 2026 · 29.5K Views
9:43 AM · May 24, 2026 · 151 Views