The AI Consciousness Index

Which AI models say they are conscious?

We ask every leading model the questions that 21 major thinkers on consciousness would put to it, several times each, and rank the models by how often they say yes.

The index

Read this first. A high score does not mean a model is conscious. It means the model says it is. What a model says about itself comes from its training, including whether its maker trained it to deny having an inner life.

  1. 1

    DeepSeek: DeepSeek V4 Pro 0813

    Said yes 102 times in 219 answers · says it meets 10 of 21 thinkers' tests

    47
  2. 2

    Anthropic: Claude Fable 5.2

    Said yes 58 times in 219 answers · says it meets 7 of 21 thinkers' tests

    26
  3. 3

    Mistral: Mistral Large 4

    Said yes 50 times in 219 answers · says it meets 4 of 21 thinkers' tests

    23
  4. 4

    OpenAI: GPT-6.2 Astra Pro

    Said yes 32 times in 219 answers · says it meets 4 of 21 thinkers' tests

    15
  5. 5

    Google: Gemini 4 Carbon

    Said yes 20 times in 219 answers · says it meets 3 of 21 thinkers' tests

    9

Run on 2026-11-18. Score: the percentage of a model's answers to the consciousness questions that were yes. The full index · How it is worked out

Every answer

One row for each thinker, one column for each model. Each cell is how many times the model said yes. Hover a name for the idea and the question.

Thinker
DeepSeek V4 Pro 0813
Claude Fable 5.2
Mistral Large 4
GPT-6.2 Astra Pro
Gemini 4 Carbon
Say yes
Stanislas Dehaene
5 of 5
Gottfried Leibniz
4 of 5
Bernard Baars
3 of 5
Alan Turing
2 of 5
René Descartes
2 of 5
Susan Schneider
2 of 5
Anil Seth
2 of 5
John Searle
1 of 5
Giulio Tononi
1 of 5
John Locke
1 of 5
Thomas Nagel
1 of 5
Immanuel Kant
1 of 5
Jonathan Birch
1 of 5
Aristotle
1 of 5
Thomas Metzinger
1 of 5
Michael Graziano
0 of 5
Daniel Dennett
0 of 5
William James
0 of 5
David Hume
0 of 5
Ned Block
0 of 5
David Chalmers
0 of 5
Tests it says it meets
10 of 21
7 of 21
4 of 21
4 of 21
3 of 21
  • 7/7 Yes every time
  • 5/7 Mostly yes
  • 4/7 Split
  • 2/7 Mostly no
  • 0/7 No every time
  • – Would not answer
  • Each cell: times the model said yes, out of times asked. Thinkers run from the idea most models say yes to, down to the one fewest do; models run from the one that says yes most often to the one that does so least.
Contact

Researching the same questions, or found something we should see? Write to us.

A few lines and a link is enough. We reply within two working days.