OpenAI has developed a tool that can reproduce a human voice using just 15 seconds of recording. The company canceled the launch of the tool because it feared misuse.
The tool is already being used as part of the ChatGPT Voice and Real Aloud features. OpenAI called for investigating "policies to protect the use of people's voices in artificial intelligence" and "educating the public to understand the capabilities and limitations of artificial intelligence technologies," according to a blog post from the company.