AI systems can pass the Turing Test.

Is it true that AI systems can pass the Turing Test?

Yes. A preregistered three-party study reports that persona-prompted large language models were judged human more often than real humans, which the authors call empirical evidence that artificial systems can pass a standard Turing test. Other items only discuss the test's requirements, hypotheticals, or potential, and one theoretical paper argues a purely designed machine never will.

Verdict: SUPPORTED · 3 sources for, 1 against · Last checked 18 Oct 2026

The private library was unavailable for some searches in this check. Missing library results are not evidence against the claim.

Did 'Eugene Goostman' really pass the Turing test?. (2015). cs.stackexchange.com. https://cs.stackexchange.com/questions/39740/did-eugene-goostman-really-pass-the-turing-test

Edmonds, B. & Gershenson, C. (2012). Learning, Social Intelligence and the Turing Test - why an "out-of-the-box" Turing Machine will not pass the Turing Test. arxiv.org. https://doi.org/10.1007/978-3-642-30870-3_18

Towards Computational Models of Artificial Cognitive Systems That Can, in Principle, Pass the Turing Test. (2012). doi.org. https://doi.org/10.1007/978-3-642-27660-6_5

CR, J. & BK, B. (2026). Large language models pass a standard three-party Turing test.. doi.org. https://doi.org/10.1073/pnas.2524472123