Viacom 18's Vats Talks AR/VR, Video AI

For Sudhanshu Vats, group CEO of Indian broadcast and video conglomerate Viacom 18, the future of video-content companies lies in the augmented, the virtual and the artificial.

In Part 1 of this Q&A, Vats gave his outlook on OTT opportunities for Indian broadcasters and content developers -- including Viacom 18's own OTT service, Voot. (See Vats on Voot: Viacom 18 Goes Agnostic.)

Now, in Part II, Vats delves deeply into higher-tech big video trends like augmented reality (AR), virtual reality (VR) and artificial intelligence (AI) -- and how they are enabling new programming and new ways of doing business for companies like Viacom 18.

Telco Transformation: Now that we have new technology and new trends in video with AR, VR and 360-degree video, how do you see video consumption and demand in India adapting and evolving?

Sudhanshu Vats: VR and AR are very exciting for a content company like ours... As a matter of fact, we are already doing a bit of experimentation in this space. We've [been] doing some VR content, particularly for our music programs... We will continue to play with this, but adoption of this will depend on, in my opinion, two factors: the quality of wearables that come in and the cost of those wearables.

India is a very price-sensitive country. When these wearables become affordable and easy to use is when we should be in a position to dial this up. In the meantime, our ability to do some niche content and our ability to do some films in this space always [is relevant]. So my answer to this would be that we will continue to experiment with this technology, we will continue to do some content from time to time, but large-scale adoption of this content will be dependent on the availability and accessibility -- particularly from a price point of view -- of wearables.

TT: What do you have in the pipeline now or in the near future in terms of AR or VR projects?

SV: We are looking at AR for some of our impact properties or [reality] formats. As far as virtual reality is concerned, we do a lot of live programming -- live entertainment -- particularly of music and so on and so forth. For instance, we do an EDM (electronic dance music) festival. We get 50,000 people into the festival, but if we can do VR shooting of this, we can give a lifelike experience to many others who are not there. That's where we are experimenting a little bit with VR.

TT: Let's talk about all of this stuff in the context of big data and analyzing viewers. How are you collecting that, and what trends are you seeing?

SV: I think we've got to be careful, consistent and compliant with the law of the land -- and make the consumer aware.

Having said that, what I'm seeing is that as long as the consumer is willingly able to share with us, and there are instances of that, we will get considerable amounts of data. And as we move forward, one of the most powerful tools we'll be able to build is our ability to look at the consumer holistically -- at least from the watch side and behavior and habits side.

What we have already seen is that, if you look at our Voot example, the kind of trust consumers are [placing] in us is of a higher order. Of our 25 million downloads, for about half of them, people have voluntarily come in as signed in. That's a good sign.

TT: What potential do you see for AI in the collection and analysis of this type of data?

SV: I see a huge potential. I'm personally a big fan of AI and machine learning.

Once we build this [consumer] database, both on the watch side and hopefully a little bit on the buy side as well, we should be able to build machine learning and AI moving forward. This is a slightly more futuristic thing for us [in the] medium term.

Work is happening now. We should be able to build recommendation engines... but we should also be able to build other areas of artificial intelligence so that that allows us to strengthen both the quality of our programming [and], more importantly, the quality of engagement and connectedness as we go forward.

TT: Do you see a potential, then, for -- if not real-time adjustments -- faster adjustments...

SV: (excited) Yes! Yes. Yes.

TT: ...in terms of the programming? Can you tell me a little about that?

SV: I see clear potential for us being able to tell stories that will not only resonate with people but can be partly crowdsourced. What I also see is the ability for us to look at stories and adapt and adjust them based on the learning we have... with a greater degree of certainty. Even today, if you look at some of the series that run in India, there is a considerable amount of adjustments that we do to our plot, to our story, to the way the story progresses and [to] certain scenarios based on the data that we get. But that data at the moment is more limited data. It is not individual-level, "deep" data. As we dial up big data, our ability to dial this piece up [by] an order of magnitude will go up considerably.

— Joe Stanganelli, Contributing Writer, Telco Transformation