TLDR: Buzz Hays, Google Cloud’s global lead for entertainment industry solutions, is actively educating Hollywood professionals on the nuances of generative AI, particularly the process of fine-tuning models. He emphasizes the importance of correct terminology and understanding the technology’s capabilities and limitations, citing Warner Bros. Discovery’s ‘The Wizard of Oz’ Sphere event as a prime example of its application.
Buzz Hays, the global lead for Google Cloud’s entertainment industry solutions, is on a mission to demystify generative AI for Hollywood. Hays notes a prevalent misunderstanding of key AI concepts, especially ‘fine-tuning’ models, and is working to clarify these terms for his colleagues in the entertainment industry. He believes a clear understanding of the technology’s pros and cons is crucial as AI becomes increasingly integrated into film and TV production.
A significant case study Hays uses to illustrate fine-tuning is the work his team undertook for Warner Bros. Discovery’s ‘The Wizard of Oz’ event at the Sphere in Las Vegas in August. This project required a high degree of fidelity to established intellectual property (IP).
Hays explained, ‘The thing that got us through this project is a process that sometimes gets confused, and I’ve heard people use the terms incorrectly, but there’s a process called fine-tuning of the models.’ He elaborated on the distinction: ‘I could say, give me an output of the Tin Man doing something without the model understanding anything about the actual Tin Man, so it comes up with some vague approximation. But that’s not good enough when you’re doing something like this.’
To achieve the desired accuracy for ‘The Wizard of Oz’ project, Google Cloud’s models were extensively fine-tuned. Hays detailed the process: ‘So we actually fine-tuned our models, which is a process that basically allows any IP owner to take a base model that’s been trained and has been using all the sources that people have opted into, to contribute to. So it’s a legitimate model, but it doesn’t understand the Tin Man. So what we can now do is fine-tune it on the Tin Man to say, ‘OK, you know how people move. Now we have a person that looks like this. So whenever we refer to Tin Man, please use this version of Tin Man for that.” This approach allowed for faithful reproductions of the main characters, regardless of their position or proximity to the camera.
Hays stressed the growing importance of this technology, particularly for franchise titles. ‘That’s where the power of AI comes. I think especially with franchise titles, we have to have a certain likeness that is persistent throughout the series, then it starts to make sense that these fine-tune models exist for that particular reason,’ he stated. He believes that the successful leveraging of these models will vary depending on the specific output requirements.
Also Read:
- Discover Financial Services and Google Cloud Partner to Revolutionize Contact Centers with Generative AI
- KPMG and Google Cloud Deepen Alliance with Major AI Investment
Despite the transformative potential, Google Cloud has implemented ‘significant guardrails’ to protect IP and individual likenesses. Hays confirmed, ‘We don’t allow our models to generate a known celebrity or likeness of anybody. There’s very careful protections put on those.’ He also highlighted other ethical safeguards: ‘We have guardrails on things like violence. We have guardrails on generating children, a bunch of these things.’ This commitment to responsible AI development aims to ensure the technology’s expansion in entertainment is both innovative and secure.
