ChatGPT deactivates a voice deemed too much like that of Scarlett Johansson | EUROtoday

Get real time updates directly on you device, subscribe now.

VShatGPT will change tone. OpenAI introduced this Monday, May 20, that it was going to droop its “Sky” voice, deemed too much like that of actress Scarlett Johansson. The latter accused the substitute intelligence firm of knowingly copying her voice, impressed by her interpretation within the movie Herthrough which Joaquin Phoenix falls in love together with his digital assistant Samantha, from whom the creators of ChatGPT have made no secret of drawing inspiration.

Evening replace

Every night from 6 p.m.

Receive the knowledge analyzed and deciphered by the Point editorial workers.

Your registration has been taken into consideration with the e-mail tackle:

To uncover all our different newsletters, go right here: MyAccount

By registering, you settle for the final circumstances of use and our confidentiality coverage.

“We have heard the questions about how we chose the voices for ChatGPT,” OpenAI mentioned on X. And clarified: “We are working to droop using Sky whereas we reply. »

“Sky's voice is in no way an imitation of that of Scarlett Johansson,” nevertheless, assured OpenAI in an article posted on its weblog, making certain that it had been developed primarily based on the voices of various actresses.

An supply declined by Scarlett Johansson

Scarlett Johansson spoke out in a press release on Monday, the place she defined that she needed to rent authorized counsel to have “Sky” faraway from the ChatGPT system, now able to talking. The actress claims to have acquired a suggestion from Sam Altam to be the voice of ChatGPT. An supply that she refused for private causes.

“Last September, I received an offer from Sam Altman to hire me to be the voice of the current ChatGPT 4.0 system. […] He said he thought my voice would comfort people,” she detailed. According to the actress, her agent was additionally contacted by the corporate, asking her to “reconsider” her refusal.

“When I heard the demo released, I was shocked, angry and in disbelief that Mr. Altman [le directeur d’Open AI, NDLR] developed a voice that sounded so eerily like mine that my closest friends and the media couldn't tell the difference,” she says.

“At a time when we are all struggling with “deepfakes” and defending our personal picture, our personal work, our personal identification, I feel these points deserve absolute readability,” she additional lamented, saying he “impatiently” waits for “appropriate legislation to help ensure the protection of individual rights” to be adopted.

On May 13, the corporate offered the brand new model of its flagship product, with GPT-4o, with improved efficiency and habits that’s supposed to be nearer to people, with new voices to return.