If we had a large dataset of experts interrogating AI/people and noting answers that raised suspicion, we'd have AI passing the Turing test more often than actu

If we had a large dataset of experts interrogating AI/people and noting answers that raised suspicion, we'd have AI passing the Turing test more often than actual people very quickly.