We all know the story of how Netflix, an online streaming platform that was once rejected by Blockbuster Video, took the world by storm and reinvented the way many of us watch our movies and TV shows.
https://www.independent.co.uk/news/world/americas/blockbuster-ceo-netflix-meeting-laugh-b2009607.html

Netflix managed to achieve this goal by utilizing the power of big data analytics and artificial intelligence to tailor its customer service. Since Netflix’s users entered into an agreement with the platform authorizing it to use their data to enhance its services, the platform had a powerful source of data with which to train its AI and ML systems.

One example of a service provided by the platform that greatly it’s enhanced user satisfaction, and helped build the platform into what it is today, is its TV show and movie recommendation system. The company famously developed a sophisticated AI system that allowed it to analyze huge amounts of data to understand which types of users liked which movies and shows, etc. This was the rocket fuel that propelled Netflix past the competition and up into the stratosphere.

While other companies like Amazon and Disney raced to catch up, Netflix became a household name that then changed the game even further when it became its own studio, producing many popular TV shows and movies.

The real problem that its rivals, as well as AI-powered TV and movie companies, had was that Netflix refused to share the source of its success, namely, its huge troves of data. This, in effect, stunted the growth of these companies as they were limited by the amount and quality of data that they could train their AI systems with.

But by hoarding its data, is Netflix doing itself more harm than good? Let’s take a look.

Open Source Data

Some of you might have heard that Tesla CEO Elon Musk famously insists that the company’s new inventions be open source, and therefore, freely available to all. His reasoning is that any company that stands still is doomed to failure, and so by sharing its patents, Tesla is incentivizing itself to work even harder to innovate.

However, there is also another cunning advantage to Musk’s strategy, namely that it encourages its rivals to rely on parts designed by Tesla, thereby stifling their own innovation. While the largest companies like Toyota and GM do have the resources to hire the top engineers, who can look at Tesla’s designs and have the time and resources to try to improve them in a manner that might possibly help them to get out front, the vast majority of car makers, particularly the numerous small electric vehicle startups, don’t.

So what if we apply this logic to Netflix and other large video-on-demand platforms? Let’s just imagine for a second that Netflix was to allow its data to be open source for everyone to use.

Well, let’s just start by saying that there are strong reasons why companies like these don’t. The main disadvantage is that your direct rivals can learn a lot from your data while still hoarding their own, something which gives them a competitive edge. But could there be advantages to sharing their data that outweighs the damage done by their rivals using it while not sharing their own?

Private companies that are involved in applying AI to media-related industry segments have sprung up everywhere. These companies have developed their own AI systems to solve specific problems ranging from helping filmmakers to improve their movies to restoring damaged silent films to allow new generations of moviegoers to see them.

As a result, all of these companies have developed unique AI systems that have gained a level of expertise that makes them extremely valuable indeed. Apple recently made massive headlines with the release of its M1 Ultra chip. This chip put the company’s hardware way out in front of its rivals like Samsung. However, it was only possible to develop thanks to Apple buying a private chipmaking company several years before, one of the hundreds of acquisitions the company has made in the last decade.

This would certainly be an option for Netflix, but an expensive one, and perhaps not the best option for a company that has begun to lose subscribers after a series of price rises to fund all the original content the company has been churning out over the past years.

Were Netflix to make its data open source then it would benefit from a wide range of companies doing the hard work for it. It would not have to worry about losing money on companies which it had bought but doesn’t deliver either.

Let’s take the example of the AI-assisted moviemaking industry. This industry segment focuses on developing and providing filmmakers an ever-widening range of tools that give them insights into what their potential audiences will think of their movies.

All of these companies are working flat-out to develop and refine new tools, all they need to make this happen is as much accurate data as possible. Were Netflix to provide this, it would dramatically boost the development and accuracy of these tools when it comes to video streaming trends and preferences, etc.

Netflix, would in turn, then be able to benefit by doing none of the hard work but still being able to access these services, admittedly, for a small fee. A good example of why this would be important is to consider international markets. Netflix is present in regions all around the world, and so collects data on all these different types of users.

However, locally developed AI systems are likely to be trained in more accurate ways given that their engineers and data scientists are local people who understand the unique cultures of their particular regions. While Netflix could employ staff in these regions to overcome this, doing this for every niche AI system would be immensely costly.

Another example are the companies that are developing tools to apply facial recognition AI to allow more detailed analysis of film and TV scenes. In this example, consider what a difference having Netflix’s data on which scenes users had turned the video off during because they were too scary, etc., would make to these companies when teaching their AI systems to understand audiences’ responses to specific scenes. The system would be able to analyze facial responses and other factors to determine what precisely some viewers found so shocking, etc.
These are just a few examples of the immense positives that video-on-demand platforms could benefit from were they to share their data.
Netflix, like Apple and so many other companies before it, is certain to benefit from a more healthy marketplace of sophisticated AI systems that can quickly and more accurately help it solve everything from production challenges to how to create better products for regional markets.

Is it time that Netflix makes its data open-source? We shall have to wait and see.