Legal filings or major news outlets

OpenAI Develops AI Text Detection Tool with 99.9% Accuracy, Faces User Backlash Concerns

Aug 5, 2024, 05:49 AM

OpenAI is developing a tool to detect and mark AI-generated text, which could significantly impact user experience and content authenticity. This watermarking technology aims to prevent misuse of AI-generated content but raises concerns about potential false accusations of AI use. According to a survey by OpenAI, 69% of ChatGPT users believe that cheating detection technology could lead to false accusations, and nearly 30% said they would use ChatGPT less if it deployed watermarks and a rival did not. OpenAI's internal AI writing detector boasts an accuracy of 99.9%, yet it has not been released for nearly two years, as reported by WSJ. The company faces a dilemma between maintaining user experience and ensuring ethical transparency.