Back to today’s briefing

Story perspectives

ChatGPT-4 Matches Six-Year-Old's Social Reasoning Skills

1/14/2025

33 3

Summary

Story summary

Recent research published in the Proceedings of the National Academy of Sciences shows that ChatGPT-4 successfully tackles 75% of theory of mind tasks, paralleling the capabilities of a six-year-old. This breakthrough underscores the remarkable progress in AI's social reasoning, igniting discussions about the nature of consciousness and the broader implications for our understanding of intelligence.

1 of 1

Story summary

Recent research published in the Proceedings of the National Academy of Sciences shows that ChatGPT-4 successfully tackles 75% of theory of mind tasks, paralleling the capabilities of a six-year-old. This breakthrough underscores the remarkable progress in AI's social reasoning, igniting discussions about the nature of consciousness and the broader implications for our understanding of intelligence.