Read time: 6 minutes

The Controversy 

Scarlett Johansson recently accused OpenAI of using her voice without permission for their ChatGPT’s “Sky” voice. This stems from CEO Sam Altman reaching out to Johansson to license her voice, which she declined. Altman's cryptic tweet, "her," didn't help matters, making people think of Johansson’s role in the movie "Her."

To review: Johansson starred in the 2013 movie "Her," where she provided the voice of Samantha, a chatbot so charming and useful that one of its users, played by Joaquin Phoenix, couldn’t help but fall in love with it. Samantha approximates the vision that OpenAI has for its own chatbots, which it claims are careening toward superhuman intelligence faster than it ever imagined.

We've managed to obtain the *actual transcript of the phone call between Sam Altman and Scarlett Johansson:

Sam: Hey Scarlett, it's Sam here.

Scarlett: Erm, Sam who?

Sam: Sam Altman, you've probably heard of m....

Scarlett: Nope I haven't actually, how did you get my number? 

Sam: Oh, uh, well, our neural networks have analysed, like, all human-generated data, and I guess your number was in there somewhere. We're launching a new update and we think if we could use your voice we'll make more money...I mean, we'll bridge the gap between tech and creatives!

Scarlett: I don't think so Sam, who are you again?

Sam: You have a very attractive voice. Do you want to chat about the singularity?

Scarlett: Please don't ever call me again.

*Possibly not the actual transcript. 

The Reality

OpenAI maintains they hired a different actress before contacting Johansson. This actress, who wished to remain anonymous, confirmed that her natural voice sounds eerily similar to Johansson's. OpenAI emphasised that they never intended to mimic Johansson.

Historical Precedent: Ford v. Bette Midler

OpenAI’s recent controversy with Scarlett Johansson brings to mind a landmark case from the 1980s: Ford v. Bette Midler. This case set a significant precedent when Bette Midler sued Ford Motor Company for using an impersonator’s voice in their advertisements after she refused to lend her own. Midler won the case, and it was established that a person’s voice could be protected against unauthorised commercial use.

The similarities between the two cases are striking. Johansson’s claim, much like Midler’s, revolves around the unauthorised use of a distinctive voice. Legal experts suggest that Johansson may have a strong case if she decides to pursue legal action. On the other hand, some argue that OpenAI’s case is different since they didn’t use her actual recordings and had already cast another actress.

Legal Perspectives

Legal experts are divided. Some argue Johansson has a strong case, similar to Midler's. Others believe it’s less clear-cut since OpenAI didn’t explicitly use her recordings. Altman's communications with Johansson raise questions about OpenAI's intent.

OpenAI’s Position

In response to Johansson’s allegations, OpenAI halted the use of the Sky voice and provided a detailed account of their voice selection process. According to OpenAI, the selection of the actress who voiced Sky was made long before Johansson was approached. They highlighted that the Sky voice was meant to be warm, engaging, and charismatic, not a clone of Johansson’s.

OpenAI’s transparency in handling the situation is commendable. They published a blog post detailing the development of five different AI voices, aiming to provide a clear picture of their intentions and processes. This move was intended to assure the public and Johansson that there was no intent to mimic her voice specifically.

The Broader Implications

Johansson’s case underscores the broader issues of AI and intellectual property. As AI technology evolves, the line between inspiration and imitation can become blurred. The legal framework surrounding AI-generated content is still in its infancy, and cases like Johansson’s will likely shape future regulations.

The debate also brings to light the ethical considerations of AI development. Companies must navigate these challenges carefully, ensuring that they respect individual rights while pushing the boundaries of innovation. Transparency and ethical standards should be at the forefront of AI development, guiding companies to make responsible decisions.

Conclusion

While OpenAI insists there was no wrongdoing, the resemblance of Sky’s voice to Johansson’s raises valid concerns about AI’s impact on intellectual property. This case serves as a reminder of the need for clear guidelines and ethical standards in AI development.

By understanding the dynamics of such cases, companies can better navigate the complex landscape of AI and intellectual property, ensuring ethical practices while driving innovation.

The tech world must balance innovation with respect for individual rights, paving the way for a future where AI enhances human creativity without infringing upon it.

Want to Implement AI within your Business? 

If you want to implement AI and ensure you're doing so with a full understanding of copyright, IP and regulation then get in touch. We partner with legal experts to ensure our clients are fully in the know.

Take the first step today and explore how AI can transform the way you work - let’s chat.