AI text detectors tend to flag text from non-native speakers as AI generated

In a study recently published in the journal Patterns, researchers demonstrate that computer algorithms often used to identify AI-generated text frequently falsely label articles written by non-native language speakers as being created by artificial intelligence. The researchers warn that the unreliable performance of these AI text-detection programs could adversely affect many individuals, including students and job applicants.