The paper proposes an auditory Turing test of 917 challenges and reports that the best tested AI model, GPT-4o audio, scores 6.9% versus 52% for a nine-person human sample.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Moravec's Paradox: Towards an Auditory Turing Test
The paper proposes an auditory Turing test of 917 challenges and reports that the best tested AI model, GPT-4o audio, scores 6.9% versus 52% for a nine-person human sample.