OpenAI has introduced its Voice Engine technology, which can clone a person’s voice using just 15 seconds of recording, Associated Press reports.
However, the company has decided not to release the technology publicly due to safety concerns. OpenAI plans to offer previews to early testers but will refrain from widespread release due to concerns about misuse, especially during an election year. While several startups already offer voice-cloning technology, OpenAI’s approach involves early testers agreeing not to impersonate individuals without consent and to disclose that the voices are AI-generated.
This decision comes in the wake of investigations into AI-generated robocalls mimicking US President Joe Biden’s voice during the New Hampshire presidential primary.
