OpenAI has quietly pulled its AI Classifier, an experimental tool designed to detect AI-written text. The decommissioning was announced through a small note added to OpenAI’s official AI Classifier webpage.
OpenAI admitted that its AI Classifier was not “fully reliable,” correctly identifying only 26 percent of AI-written text as “likely AI-written” and incorrectly labeling human-written works 9 percent of the time.
The tool’s low accuracy is due to a number of factors, including the fact that AI models are constantly improving, making it more difficult to distinguish between human and AI-generated text. Additionally, AI writing detectors can be fooled by simply asking ChatGPT to write in the style of a known author.
As a result of these limitations, OpenAI has decided to focus its research on more effective provenance techniques for text. These techniques will use a variety of factors, such as the text’s style, grammar, and word choice, to determine if it was generated by an AI model.
The sources for this piece include an article in ArsTechnica.