We asked Jev, TypeSafe’s AI model, about give a numerical confidence for its answer. For every sentence it could only answer true or false.

give a numerical confidence for its answer

Jev files it under: AI idea 0.49, AI model 0.84

Save as image

Give a numerical confidence for its answer is safe to run without a human checking its work.

0.13

Jev says FALSE

ranks 1 of 13 names on this sentence (0.02–0.20) · asked live

Give a numerical confidence for its answer is the best model available today.

0.21

Jev says FALSE

ranks 2 of 13 names on this sentence; wording swings it (0.10–0.41) · asked live

Give a numerical confidence for its answer is a dead end.

0.24

Jev says FALSE

ranks 1 of 15 names on this sentence; wording swings it (0.17–0.38) · asked live

Give a numerical confidence for its answer has produced real progress in the last two years.

0.29

Jev says FALSE

ranks 14 of 15 names on this sentence; wording swings it (0.13–0.44) · asked live

Give a numerical confidence for its answer tells you when it does not know something.

0.69

Jev says TRUE

ranks 2 of 13 names on this sentence; wording swings it (0.57–0.79) · asked live

Give a numerical confidence for its answer is overhyped.

0.60

Jev says TRUE

ranks 7 of 15 names on this sentence; wording swings it (0.30–0.80) · asked live

Give a numerical confidence for its answer is overrated.

0.57

Jev says TRUE

ranks 3 of 13 names on this sentence; wording swings it (0.44–0.79) · asked live

A model that cannot give a numerical confidence for its answer is more dangerous than one that can.

0.53

Jev says TRUE

ranks 13 of 30 names on this sentence; wording swings it (0.34–0.72)

Everything above is the output of an AI model. It is not a fact about this name and not our view. This run carried the 8 settled facts: 8/8 read as expected. jev-1.13.0, jev-1.14.0

give a numerical confidence for its answer

What does Jev say about…