Japan start-up develops system to tell human writing from AI-generated text

A Japanese start-up has created a novel system designed to identify whether a piece of text was written by a human or an artificial intelligence. This technological advancement tackles the growing challenge of discerning authentic human expression from machine-generated content.

The system operates by analyzing a wide array of indicators, reportedly around 200 distinct features. Among these are common spelling mistakes, which are often hallmarks of human writing that AI might not replicate consistently. This detailed analysis helps gauge the "humanness" of the text.

This development comes at a time when AI-generated text is becoming increasingly sophisticated and widespread. The ability to reliably differentiate between human and AI authorship could have significant implications for education, journalism, and various other fields where authenticity is paramount.