OpenAI Pauses ChatGPT-4o Voice Amid Accusations of Scarlett Johansson Imitation

In a move that highlights the growing ethical and legal challenges surrounding AI voice mimicry, OpenAI has temporarily paused the "Sky" voice mode for its latest chatbot, ChatGPT-4o. The decision comes after widespread backlash from fans who accused the company of deliberately copying actress Scarlett Johansson's iconic performance in the 2013 sci-fi film Her.

The controversy began when users noticed that Sky’s voice bore a striking resemblance to Johansson’s character, Samantha, a digital assistant with whom the protagonist forms an intimate relationship. The similarity was so uncanny that even close friends and family members of Johansson were confused, leading to widespread speculation about the AI company's intentions.

OpenAI attempted to address the concerns in a detailed blog post explaining their process for selecting voices for ChatGPT-4o. According to the company, Sky’s voice is not an imitation of Johansson but rather belongs to a different professional actress using her natural speaking voice. However, OpenAI acknowledged that the resemblance was too close to ignore and decided to pause the feature.

“We believe that AI voices should not deliberately mimic a celebrity’s distinctive voice,” OpenAI stated in their blog. “Sky’s voice is not an imitation of Scarlett Johansson but belongs to a different professional actress using her own natural speaking voice.”

The company has chosen to remain tight-lipped about the identity of the voice actors, citing privacy concerns. This decision, however, did little to quell the public's skepticism and outrage.

The situation took a dramatic turn when Johansson herself released a statement addressing the issue. In her message, she revealed that OpenAI CEO Sam Altman had approached her last September with an offer to voice the ChatGPT 4.0 system. According to Johansson, Altman believed that her involvement could help bridge the gap between tech companies and creatives, making AI more accessible and comfortable for users.

“After much consideration and for personal reasons, I declined the offer,” Johansson wrote. “Nine months later, my friends, family, and the general public all noted how much the newest system named ‘Sky’ sounded like me.”

Johansson expressed shock and anger upon hearing the demo of Sky’s voice, which she described as eerily similar to her own. She also pointed out that Altman had tweeted a single word, “her,” seemingly referencing the film Her and implying an intentional mimicry.

“Two days before the ChatGPT 4.0 demo was released, Mr. Altman contacted my agent, asking me to reconsider,” Johansson continued. “Before we could connect, the system was out there.”

Faced with this situation, Johansson hired legal counsel, who sent two letters to Altman and OpenAI, demanding an explanation of how the Sky voice was created. As a result, OpenAI reluctantly agreed to take down the Sky voice.

This incident underscores the complex ethical and legal landscape surrounding AI-generated voices. While AI has the potential to revolutionize various industries, it also raises significant concerns about consent, intellectual property, and the rights of individuals whose voices or likenesses are used without their explicit permission.

As AI continues to evolve, policymakers and tech companies will need to navigate these challenges carefully to ensure that innovation does not come at the expense of individual rights and privacy. For now, OpenAI’s decision to pause Sky’s voice is a step towards addressing these concerns and rebuilding public trust.