Hollywood actress and Oscar nominee Scarlett Johansson has voiced her outrage over the alleged unauthorized use of her voice by tech giant OpenAI. Johansson expressed shock and disbelief that the company’s CEO, Sam Altman, would go to such lengths to replicate her voice for the latest ChatGPT 4.0 system without her consent.
In a report by Sky News, Johansson detailed how she had previously declined Altman’s offer to lend her voice to the AI system for personal reasons. Despite this, she later discovered that the voice used in ChatGPT 4.0 sounded strikingly similar to hers.
Recalling her role in the 2013 sci-fi film ‘Her,’ where she voiced an AI named Samantha, Johansson noted the eerie resemblance between her character and the AI voice named “Sky” in the new ChatGPT system.
“Last September, Sam Altman approached me to voice ChatGPT 4.0, suggesting that my voice could bridge the gap between tech and creativity, and help ease the public’s concerns about AI,” Johansson explained. “After careful consideration, I declined the offer. Nine months later, I was shocked to find out that the system’s voice was almost identical to mine.”
Johansson described her frustration upon hearing the demo, which left her feeling “shocked, angered, and in disbelief.” She added that Altman had seemingly acknowledged the resemblance by tweeting “her,” a direct reference to her role in ‘Her.’
Before the demo’s release, Altman reportedly reached out to Johansson’s agent to reconsider the offer, but the system went live before they could connect. This led Johansson to hire legal counsel, who wrote to OpenAI demanding an explanation of how they created the “Sky” voice. Subsequently, OpenAI agreed to remove the “Sky” voice.
The incident highlights growing concerns over the ethical use of AI, particularly in replicating individuals’ voices and images without consent. Johansson emphasized the importance of addressing these issues, stating, “In an era of deepfakes and the need to protect our identities and work, these questions demand absolute clarity. I hope for transparency and appropriate legislation to safeguard individual rights.”
In response, OpenAI tweeted that it had “paused” the use of the “Sky” voice while addressing the concerns raised. They clarified in a statement that the voices, including “Sky,” were created in collaboration with voice actors.
The controversy follows previous legal challenges against OpenAI, such as the 2023 lawsuit by the New York Times, which alleged that OpenAI and its major investor, Microsoft, unlawfully used the newspaper’s articles to train ChatGPT.
This case underscores the pressing need for ethical guidelines and legal frameworks to govern the use of AI in replicating human likenesses, ensuring respect for personal rights and consent.
