Three Ways AI will Solve Hollywood’s Data Problem
AI was present in every corner of NAB Show this year, and for a good reason. The rapid adoption of AI and Machine Learning may solve some of the media industry’s largest problems. But before they can do that, the industry needs to understand what the problem is and what exactly AI can do.
I sat down with data scientist Yves Bergquist, the founder and director of the Data & Analytics Project at University of Southern California’s Entertainment Technology Center and CEO of AI engineering firm Novamente, to talk about AI and how it will change the future of filmmaking. Here’s his perspective:
What’s the Difference Between Machine Learning and Artificial Intelligence?
YB: There is often a lot of confusion about the difference between Machine Learning and AI. Many things are labeled AI that aren’t AI at all. Let me try to help differentiate.
Artificial Intelligence is the ability to represent complex data and reason from it. Here’s a great example: You are running with a bucket full of nails. Likely, you’ve never actually run with a bucket full of nails. However, you know what a bucket is, you’ve run before and you’ve held a container filled with something. Now your brain is doing something amazing! It is picking from related experiences and inferring what this experience will be like. That is what AI is trying to do.
Simply put, the difference between Machine Learning and AI is:
- Machine Learning = learning machines
- Artificial Intelligence = thinking machines.
Our ability to learn as humans is a foundation of our intelligence. Similarly, AI is a broader area that includes a dozen or so domains that are each key to the development of thinking machines; the ability to learn is just one component.
What is Hollywood’s Data Problem and How Will AI Solve It?
YB: Media and entertainment has several obstacles when it comes to how it uses data, or rather its lag on collecting and leveraging data for insights.
- Audience Intelligence
The film industry is essentially a B2B business. They don’t have access to nor collect much data about their end-consumer. Not even the movie theaters have much intelligence about viewers. Sure, they may know how many people came and how much money they spent, but they don’t know how old they are, their gender or why they came to watch a particular film. Currently, the film industry is making products for consumers they have very little granular data about, and their product is very, very expensive to make. You can see why they are incredibly eager to find new ways to collect more data and interpret it, and AI can take data that is complex, ambiguous and sparse (social media and other noise) and deliver new insights.
- Knowledge Representation
Another key challenge is that companies have a lot of data in many different formats and structures. It requires a lot of human ‘massaging’ and normalization to make sense of it. Now the media industry has this problem on steroids! There’s video files, production notes, social media noise, numerical data about box office and marketing spend, etc. It has data in arguably hundreds of different formats and it would take hundreds of data scientists to extract insights from it. Knowledge representation will be a huge area of growth for AI, and for media and entertainment. Expect to see all of this data put into a container that can automatically and autonomously connect the dots without needing too much human interaction.
- Narrative Algorithms
Audiences have a powerful relationship with film stories, but it’s never been possible to quantitatively measure that relationship – what’s a ‘good’ story? How should a character develop?
I recently wrote about how Hollywood and Silicon Valley are in the same business: producing algorithms. What I mean by this is that just like software algorithms, films are narratives the compress our infinite reality into a simple script. As the market evolves, the media and entertainment industry will need to start thinking algorithmically about stories and AI is shedding light on story and character mechanics that can drive box office success.
What Will Come After AI?
YB: I see the next domain we should approach is the neuroscience behind our media and entertainment choices – why do we like certain stories and dislike others? Why do we entertain ourselves? Why do we go see a movie? I hope we can apply a lot more science to what has been a non-linear field.
Bringing Data to Life
Learn how to manage your assets in a hybrid architecture:
- Simplify collaboration across geo locations
- Achieve extreme data durability and integrity
- Integrate existing environments
- Build for cloud-scale
Yves Bergquist is a data scientist and founder and director of the “Data & Analytics” Project at University of Southern California’s Entertainment Technology Center, where his team helps the entertainment industry accelerate the deployment of next-generation audience intelligence standards and solutions.
Yves is also CEO of AI engineering firm Novamente, which builds neural-symbolic cognitive architectures to solve large-scale enterprise problems.
Certain blog and other posts on this website may contain forward-looking statements, including statements relating to expectations for our product portfolio, the market for our products, product development efforts, and the capacities, capabilities and applications of our products. These forward-looking statements are subject to risks and uncertainties that could cause actual results to differ materially from those expressed in the forward-looking statements, including development challenges or delays, supply chain and logistics issues, changes in markets, demand, global economic conditions and other risks and uncertainties listed in Western Digital Corporation’s most recent quarterly and annual reports filed with the Securities and Exchange Commission, to which your attention is directed. Readers are cautioned not to place undue reliance on these forward-looking statements and we undertake no obligation to update these forward-looking statements to reflect subsequent events or circumstances.