OpenAI removes voice from ChatGPT after controversy with Scarlett Johansson

Ethical concerns raised after OpenAI boss references 2013 movie in X post

Life

Image: Warner Bros.

OpenAI has removed the voice ‘Sky’ from ChatGPT’ s voice mode following allegations of plagiarism of actress Scarlett Johansson’s voice. The controversy surrounding the voice, which was strikingly similar to Johansson’s, has raised questions about the ethics of using celebrity voices in artificial intelligence.

Users noted that the voice Sky was suspiciously similar to that of actress Scarlett Johansson. This led to speculation that OpenAI had used Johansson’s voice without her permission. Initially, OpenAI denied, but eventually removed the voice.

OpenAI first announced that the voice had nothing at all to do with that of the actress. But then Johansson revealed that she had indeed been approached by OpenAI CEO Sam Altman to voice the system, but had declined the offer. Altman also did the company no favours by tweeting a single word before ChatGPT-4o launch: ‘her’ – clearly a reference to the 2013 film directed by Spike Jonze where Johansson voiced an AI that becomes involved with a vulnerable and depressed man going through a divorce.

After legal action from Johansson’s lawyers, OpenAI decided to remove the Sky voice. The company has yet to comment on whether or not the voice will be reactivated in the future.

“I think it is important that we are open and honest about how we develop and use AI,” Johansson said. “We need to make sure that this technology is used responsibly, with respect for the rights of individuals.”

The issue calls into question the limits of AI technology and the responsibility of companies developing it. Future discussions on AI ethics will undoubtedly delve deeper into how we deal with identity, representation and copyright in the digital age.

Ethical considerations aside it’s arguable picking Johansson’s voice was not an appropriate choice to showcase the capabilities of AI. The movie her (lower case intended) has a clear message: technology is not compatible with humans.

News Wires

Read More: ChatGPT-4o OpenAI