Fake News Detectors are Biased against Texts Generated by Large Language Models

Open in new window