Google and Meta Platforms have held discussions with major Hollywood studios about licensing content for use in the tech giants’ artificial intelligence video generation software, according to people familiar with the matter.
Both companies are developing technology that can create realistic scenes from a text prompt, and have offered tens of millions of dollars to partner with studios in some capacity. Rival OpenAI, backed by Microsoft, is having similar conversations. Google, Meta and OpenAI declined to comment on the talks.
Hollywood studios are keen to discuss ways to use AI to reduce costs while also protecting themselves from having their work stolen. They are wary of giving films and TV shows to tech companies without control over how that content is used. Just this week, the actress Scarlett Johansson demanded OpenAI stop using a voice that sounded like her to power its chatbot after she refused to work with the company.
Big money is at stake. On Wednesday, News Corp, parent of the Wall Street Journal and other media outlets, agreed to allow OpenAI to use content from more than a dozen of its publications in a deal that could be worth more than US$250-million over five years.
Warner Bros Discovery has expressed a willingness to license some of its programmes to train the models, but only for specific divisions — not its entire library. Walt Disney and Netflix aren’t willing to license their content to these companies but have expressed an interest in other types of collaborations.
Hollywood studios are already using AI in production, as are many filmmakers. Tyler Perry has used the technology to recreate the makeup he wears for his Madea character in movies. Director Robert Zemeckis has deployed AI to de-age Tom Hanks in an upcoming film.
New crop of tools
But a new crop of tools, including OpenAI’s Sora and Google’s Veo, go a step further by promising to help filmmakers quickly create vivid, hyper-realistic clips based on just a few words of description. Their capabilities have elicited excitement and anxiety in Hollywood, where actors and writers staged a months-long strike last year over concern about how AI could take their jobs.
Perry, an actor, filmmaker and studio owner, was so amazed by a demonstration of Sora that he put plans for an $800-million studio expansion on hold earlier this year. He has been vocal about the opportunities AI represents for studios, but also in raising the alarm about the technology’s impact on labour. He has called for the industry to rally together and formulate some sort of regulations. “If not, I just don’t see how we survive,” he told the Hollywood Reporter in February.
The music industry has adopted a tough stance against AI use. Universal Music Group sued Anthropic, a budding AI start-up, for copying song lyrics and temporarily pulled its music from TikTok in part to secure protections for its artists. Sony Music Group sent letters to hundreds of partners this month warning them not to train any AI models on its music.
No major studio has so far sued a tech company over AI, despite fears that many of these models have already been trained on their copyrighted material. They would like to find a way to make AI work for them rather than fight a new technology that could help significantly reshape the industry. But studios have yet to agree to major commercial relationships around the use of AI with the largest technology companies.
Hollywood executives worry that licensing conversations will lead to tension between studios and their creative partners. For example, studios believe they have the rights to license a movie they own to an AI company. But if that company uses the movie to train its AI model on the face or voice of an actor in that movie, the actor would also want the opportunity to approve of it or not. A few actors have already struck deals with AI companies. — (c) 2024 Bloomberg LP