Last September, I was approached by Sam Altman with an offer to voice the ChatGPT 4.0 system. He believed that my voice could help bridge the gap between tech companies and creatives, making consumers more comfortable with the growing presence of AI in our lives. However, after careful consideration for personal reasons, I decided to decline the offer.
Nine months later, I was surprised to discover that the new system, named “Sky,” sounded remarkably similar to me. Upon hearing the demo, I was shocked and angered by the uncanny resemblance. Mr. Altman even hinted at the intentional similarity by referencing the movie “Her,” in which I voiced a chat system named Samantha who develops a close relationship with a human.
Just two days before the release of the ChatGPT 4.0 demo, Mr. Altman reached out to my agent, asking me to reconsider. However, by then, the system was already out in the public domain. As a result, I had to seek legal counsel, who sent letters to Mr. Altman and OpenAI outlining their actions and demanding an explanation of how the “Sky” voice was created. Eventually, OpenAI agreed to remove the voice.
In a time where concerns about deepfakes and the protection of personal likeness and identity are prevalent, clarity and transparency are essential. I hope for a resolution that includes transparency and the implementation of appropriate legislation to safeguard individual rights.
