After telegraphing the move in media appearances, OpenAI has launched a tool that attempts to distinguish between human-written and AI-generated text — like the text produced by the company’s own ChatGPT and GPT-3 models. The classifier isn’t particularly accurate — its success rate is around 26%, OpenAI notes — but OpenAI argues that it, when used in tandem with other methods, could be useful in helping prevent AI text generators from being abused.
“The classifier aims to help mitigate false claims that AI-generated text was written by a human. However, it still has a number of limitations — so it should be used as a complement to other methods of determining the source of text instead of being the primary decision-making tool,” an OpenAI spokesperson told TechCrunch via email. “We’re making this initial classifier available to get feedback on whether tools like this are useful, and hope to share improved methods in the …
