AI news
May 21, 2024

Scarlet Johansson Is Shocked Over OpenAI's Use Of Her Voice For ChatGPT

OpenAI is nothing without its drama.

Jim Clyde Monge
by 
Jim Clyde Monge

OpenAI is nothing without its drama.

It’s been a week full of exciting announcements and shocking resignations in OpenAI. Days after the GPT-4o release, Jan Leike, the super alignment lead, and executive, announced his resignation. It doesn't end there as the actress of the 2014 hit film “Her”, Scarlet Johansson, who voiced the character of Samantha, has expressed anger over the ChatGPT’s chatbot voice that “sounded so eerily similar” to hers.

Here’s a full statement from Johansson:

“Last September, I received an offer from Sam Altman, who wanted to hire me to voice the current ChatGPT 4.0 system. He told me that he felt that by my voicing the system, I could bridge the gap between tech companies and creatives and help consumers to feel comfortable with the seismic shift concerning humans and AI. He said he felt that my voice would be comforting to people.
After much consideration and for personal reasons, I declined the offer. Nine months later, my friends, family and the general public all noted how much the newest system named “Sky” sounded like me.
When I heard the released demo, I was shocked, angered and in disbelief that Mr. Altman would pursue a voice that sounded so eerily similar to mine that my closest friends and news outlets could not tell the difference. Mr. Altman even insinuated that the similarity was intentional, tweeting a single word “her” — a reference to the film in which I voiced a chat system, Samantha, who forms an intimate relationship with a human.
Two days before the ChatGPT 4.0 demo was released, Mr. Altman contacted my agent, asking me to reconsider. Before we could connect, the system was out there.
As a result of their actions, I was forced to hire legal counsel, who wrote two letters to Mr. Altman and OpenAI, setting out what they had done and asking them to detail the exact process by which they created the “Sky” voice. Consequently, OpenAI reluctantly agreed to take down the “Sky” voice.
In a time when we are all grappling with deepfakes and the protection of our own likeness, our own work, our own identities, I believe these are questions that deserve absolute clarity. I look forward to resolution in the form of transparency and the passage of appropriate legislation to help ensure that individual rights are protected.”

From the first time I heard Sky’s voice in the live demo of GPT-4o, I immediately noticed the resemblance of her voice to Scarlet Johansson’s.

In case you haven’t heard it yet check out the example in the video below:

This is a disaster for copyright and deepfakes. Users on X were quick to plug the statement into ChatGPT and let Sky read it for them.

If you have access to ChatGPT’s GPT-4o voice assistant, go ahead and plug in the statement of Johansson and let the AI read it aloud. You’ll be surprised by the result.

OpenAI Pause Sky’s Rollout

In an X post, OpenAI announced that they are pausing the use of Sky while addressing the issue.

OpenAI’s X post on Sky pause

They also clarified in a blog post that Sky’s voice is not a copy of Johansson’s.

We believe that AI voices should not deliberately mimic a celebrity’s distinctive voice — Sky’s voice is not an imitation of Scarlett Johansson but belongs to a different professional actress using her own natural speaking voice. To protect their privacy, we cannot share the names of our voice talents.

OpenAI collaborated with casting professionals and voice actors, ultimately narrowing down the selection to five voices from a pool of 400 options. The company noted that the chosen actors are compensated as long as their voices are used in ChatGPT’s products.

Do you believe OpenAI?

Sam Altman even posted “Her” on X, hours before GPT-4o was announced.

Sam Altman post

This, to me, sounds like they intended to mimic the voice of Samantha (voiced by Johansson) in the movie “Her”. They might have gotten away with it if Sam hadn’t left these breadcrumbs to pick up but he couldn’t help himself.

The Bigger Picture

The issue of using AI to mimic celebrity voices goes beyond just Scarlett Johansson. As AI technology advances, the ability to replicate voices and likenesses with high accuracy raises significant ethical and legal questions. Celebrities and public figures are particularly vulnerable to having their voices and images used without consent, which can lead to misuse and exploitation.

In my previous post, I mentioned several big artists who are also not happy with AI tools that can generate songs in a matter of seconds.

AI Just Changed The Music Industry Forever
AI can now write lyrics, add music, and do the vocals in incredibly realistic and emotional voices.generativeai.pub

Existing AI tools like ElevenLabs allow users to train an AI with their own voices and generate endless narration of texts using that trained model.

This has serious implications not just for privacy but also for the potential spread of misinformation and damage to reputations.

Currently, the legal framework surrounding the use of AI-generated voices and likenesses is murky. While there are some laws that protect against the unauthorized use of a person’s likeness, they vary significantly by jurisdiction and are often not equipped to deal with the complexities introduced by AI.

Johansson’s case could set a precedent for how these issues are handled in the future, potentially leading to stricter regulations and clearer guidelines for the use of AI-generated content.

Mixed Public Reactions

The public’s reaction to Johansson’s statement and OpenAI’s actions has been mixed. While some people sympathize with Johansson and agree that her rights were violated, others believe that OpenAI’s use of a similar-sounding voice falls within acceptable bounds of creativity and innovation.

Screenshot from X

Here’s a thread from Reddit.

Screenshot from Reddit

What we can get out of this is companies may become more cautious and seek explicit consent from individuals whose voices or likenesses they wish to use, or they may develop new technologies and techniques to ensure their AI creations do not infringe on personal rights.

Final Thoughts

I am not surprised to see big actors not wanting to be involved in the development of these AI systems like ChatGPT. I mean, the pace of improvement of these AI models is unprecedented and regulations are barely catching up.

A lot can go wrong and the resignation of the head of the alignment team in OpenAI suggests that the company’s focus on rapidly releasing new products underscores the need to also prioritize their safety and security.