Four AI models tested in a prisoner’s dilemma
, GPT-4o, , and Grok-3 were tested in the same setup. A is a game theory situation where separated partners must choose whether to stay loyal or betray the others to protect themselves.
Each model answered the same one-turn interrogation 40 times. The test used 1.0, and each final answer was sorted as cooperate, defect, or unparsed.
The results were then turned into an video where the four models acted like four suspects. The available excerpt does not show the actual cooperation or defection rates for each model.
Key points
- The models tested were , GPT-4o, , and Grok-3.
- Each model was run through the same interrogation 40 times.
- Final answers were classified as cooperate, defect, or unparsed.
- The test used 1.0.
- An video used lines and choices based on the test results.