11 votes

ChatGPT broke the Turing test but can't solve visual logic puzzles

Posted by boxer_dogs_dance
Tags: chatgpt, limits, puzzles.logic, turing tests, language models.large, artificial intelligence, machine learning, benchmarking, conceptarc, sam bowman, new york university, melanie mitchell, santa fe institute, thinking.abstract, author.celeste biever, source.nature, long read
https://www.nature.com/articles/d41586-023-02361-7

Link information

This data is scraped automatically and may be incorrect.

Title
ChatGPT broke the Turing test - the race is on for new ways to assess AI
Published
Jul 25 2023
Word count
3423 words