mlsanityData to 7 Oct 2026

Does GPT-4 pass the Turing test?

arXiv 2310.20216 ↗First posted 31 Oct 2023cs.AI cs.CLcited by 37

Not on the map. Why?

Abstract (arXiv)

We evaluated GPT-4 in a public online Turing test. The best-performing GPT-4 prompt passed in 49.7% of games, outperforming ELIZA (22%) and GPT-3.5 (20%), but falling short of the baseline set by human participants (66%). Participants' decisions were based mainly on linguistic style (35%) and socioemotional traits (27%), supporting the idea that intelligence, narrowly conceived, is not sufficient to pass the Turing test. Participant knowledge about LLMs and number of games played positively correlated with accuracy in detecting AI, suggesting learning and practice as possible strategies to mitigate deception. Despite known limitations as a test of intelligence, we argue that the Turing test continues to be relevant as an assessment of naturalistic communication and deception. AI models with the ability to masquerade as humans could have widespread societal consequences, and we analyse the effectiveness of different strategies and criteria for judging humanlikeness.

Key facts. Cited by 37 papers in the map (data through 7 Oct 2026). Builds on 7 of its 33 references that are in the map. First posted on arXiv 31 Oct 2023.

Builds on

7 of 33 references.

References · 7

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2019SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systemscited by 1,718
  3. 2021Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigmcited by 599
  4. 2019Defending Against Neural Fake Newscited by 553
  5. 2021AI and the Everything in the Whole Wide World Benchmarkcited by 340
  6. 2022The Alignment Problem from a Deep Learning Perspectivecited by 246
  7. 2023Language Model Behavior: A Comprehensive Surveycited by 92

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 17; 2025: 15; 2026: 5.

37 papers.

Sort:

By area:

  1. 2024Large Language Models Assume People are More Rational than We Really arecited by 49
  2. 2025Large Language Models Pass the Turing Testcited by 48
  3. 2024People cannot distinguish GPT-4 from a human in a Turing testcited by 46
  4. 2024TRAP: Targeted Random Adversarial Prompt Honeypot for Black-Box Identificationcited by 33
  5. 2024How Well Can LLMs Echo Us? Evaluating AI Chatbots' Role-Play Ability with ECHOcited by 19
  6. 2025Common Corpus: The Largest Collection of Ethical Data for LLM Pre-Trainingcited by 19
  7. 2024Lies, Damned Lies, and Distributional Language Statistics: Persuasion and Deception with Large Language Modelscited by 14
  8. 2025Thinking beyond the anthropomorphic paradigm benefits LLM researchcited by 13
  9. 2025TwinVoice: A Multi-dimensional Benchmark Towards Digital Twins via LLM Persona Simulationcited by 13
  10. 2024Helmsman of the Masses? Evaluate the Opinion Leadership of Large Language Models in the Werewolf Gamecited by 12
  11. 2024The Moral Turing Test: Evaluating Human-LLM Alignment in Moral Decision-Makingcited by 12
  12. 2024The Road to Artificial SuperIntelligence: A Comprehensive Survey of Superalignmentcited by 11
25 more
  1. 2025IMPersona: Evaluating Individual Level LM Impersonationcited by 11
  2. 2024SRAP-Agent: Simulating and Optimizing Scarce Resource Allocation Policy with LLM-based Agentcited by 8
  3. 2024GPT-4 is judged more human than humans in displaced and inverted Turing testscited by 7
  4. 2025Different types of syntactic agreement recruit the same units within large language modelscited by 7
  5. 2025Digital Player: Evaluating Large Language Models based Human-like Agent in Gamescited by 6
  6. 2024Philosophy of Cognitive Science in the Age of Deep Learningcited by 5
  7. 2024Are You Human? An Adversarial Benchmark to Expose LLMscited by 5
  8. 2025Revealing emergent human-like conceptual representations from language predictioncited by 5
  9. 2024Fostering Natural Conversation in Large Language Models with NICO: a Natural Interactive COnversation datasetcited by 4
  10. 2025The State Of TTS: A Case Study with Human Fooling Ratescited by 4
  11. 2024Dynamic Normativity: Necessary and Sufficient Conditions for Value Alignmentcited by 3
  12. 2025Assessing LLMs in Art Contexts: Critique Generation and Theory of Mind Evaluationcited by 3
  13. 2025Love, Lies, and Language Models: Investigating AI's Role in Romance-Baiting Scamscited by 3
  14. 2025A Rigorous Turing Test: a Foundation for Evaluating Artificial General Intelligencecited by 2
  15. 2026How Open Must Language Models be to Enable Reliable Scientific Inference?cited by 2
  16. 2024The Voice: Lessons on Trustworthy Conversational Agents from "Dune"cited by 1
  17. 2024Advancements and limitations of LLMs in replicating human color-word associationscited by 1
  18. 2025ChatGPT-4 and other LLMs in the Turing Test: A Critical Analysiscited by 1
  19. 2025"I Hadn't Thought About That": Creators of Human-like AI Weigh in on Ethics And Neurodivergencecited by 1
  20. 2026Human or Machine? A Preliminary Turing Test for Speech-to-Speech Interactioncited by 1
  21. 2024AI-Driven Agents with Prompts Designed for High Agreeableness Increase the Likelihood of Being Mistaken for a Human in the Turing Testcited by 0
  22. 2025The next question after Turing's question: Introducing the Grow-AI testcited by 0
  23. 2026Book your room in the Turing Hotel! A symmetric and distributed Turing Test with multiple AIs and humanscited by 0
  24. 2026Identity from the Outside: A Conceptual Framework and Research Program for AI Personality Clonescited by 0
  25. 2026Detecting AI Impostors: How Do Middle Schoolers Identify LLM Agents in a Live Collaborative Setting?cited by 0