ChatGPT suspends Scarlett Johansson-like voice amid actor’s outcry

OpenAI discontinued a voice option in its ChatGPT product that closely resembled Scarlett Johansson’s voice after the actor raised objections. Johansson expressed surprise and anger at having used a voice eerily similar to her own without her consent.

Johansson said in a statement that OpenAI had approached her nine months earlier about its AI system, but she declined for personal reasons. Despite her refusal, she was “shocked” to hear the voice option called “Sky,” which friends and news outlets said was indistinguishable from her voice.

In response to the backlash, OpenAI removed the heavily promoted Sky voice option from ChatGPT on Monday. The company introduced Sky during a high-profile event last week, demonstrating the capabilities of the new ChatGPT-4o model. The voice immediately drew comparisons to Johansson’s AI companion character in the 2013 film Her.

To add to the controversy, OpenAI CEO Sam Altman had posted a tweet after the presentation in which he said “she,” suggesting an intentional similarity to Johansson’s character. However, OpenAI later clarified that Sky was not based on Johansson, explaining in a blog post that the voice belonged to another professional actress using her natural speaking voice. To protect their privacy, the company has decided not to reveal the names of its spokespersons.

Johansson claimed that Altman initially pitched the project to her, suggesting that her voice could help bridge the gap between tech companies and creatives and give consumers more confidence in AI. Just two days before Sky’s release, he contacted her agent again and asked her to reconsider. Johansson’s lawyers then contacted OpenAI and requested that the voice be removed.

Criticism of Sky wasn’t just limited to his resemblance to Johansson. Some commentators, including Daily Show host Desi Lydic, criticized the voice as overly flattering and gendered, with Lydic joking that it was obvious that “a man developed this technology.”

OpenAI defended its selection process for ChatGPT’s voice, saying it aimed for a voice with a “timeless” quality that would inspire trust. The company reviewed hundreds of submissions before selecting five voice options for ChatGPT, which were released in September. The selected actors then took part in recording sessions to train OpenAI’s models.

Sky’s removal coincided with significant internal turmoil at OpenAI, with several senior members of the security team resigning. A key researcher, Jan Leike, criticized the company for prioritizing “shiny products” over safety protocols. Still, Altman and co-founder Greg Brockman insisted that OpenAI would not release a product if there were security concerns.

OpenAI’s blog post emphasized collaboration with entertainment industry professionals and compensation for voice actors. The technology industry, including OpenAI, is facing intense scrutiny and legal challenges from entertainers, creators and media companies over possible copyright infringement and fears that AI could replace human workers. Major entertainment unions like Sag-Aftra have gone on strike over issues such as AI’s use of likenesses.