OpenAI has confirmed that AI writing detectors cannot reliably distinguish between content generated by AI and humans, often falsely accusing students of using AI to write their work. Despite some tools claiming to detect AI-generated content, none have proven reliable.
OpenAI also clarifies that its AI model, ChatGPT, cannot identify whether text is AI-generated, and may provide incorrect information, hence users should not use it as their sole source of research. While automated AI detectors are ineffective, humans can sometimes detect AI writing based on typical writing styles and careless errors.