Original Source
Advanced AI Passes Turing Test for First Time
AI Passes Turing Test for the First Time
A recent study has provided the first empirical proof of an advanced artificial intelligence (AI) model, GPT-4.5, passing the Turing test. This research demonstrates that AI can imitate human conversation so convincingly that people often cannot distinguish it from a real person. When equipped with specific 'persona' prompts, the AI was judged to be human 73% of the time by participants, significantly outperforming actual human participants and fundamentally altering our understanding of machine intelligence.
Implications of the Study
The University of California San Diego study, published in the Proceedings of the National Academy of Sciences, rigorously applied the 1950 framework created by British mathematician Alan Turing. This marks the first time models were judged to be human as often as actual humans using Turing's method. Dr. Cameron Jones, the study's corresponding author, noted that advanced large language models (LLMs) can exhibit the same tone, directness, humor, and fallibility as humans when given the right prompts. This finding suggests major implications for how we perceive AI, beyond its ability to produce knowledge on nearly any topic, now extending to convincing social behavioral traits.
*Source: Neuroscience News (2026-05-19)*
Related Articles
📧 Daily Newsletter
Get the daily global news briefing in your inbox every morning.
It's still free.