Generative AI and sustainability: Vital, but at a price

Harrison commenced by asking speakers to nominate a dominant theme, and Gordon Brooks, Executive Chairman and CEO of Zixi, picked Chat GPT for how it might be used in content creation.

He said: “It is not to do the actual content creation but to give people ideas. I though it was interesting too that sustainability stayed on the list of predicted trends, but it comes with a caveat this year. It has to be cost justified.”

Caroline Brown, an Independent with work experience at Comcast, Warner Brothers and Discovery on her CV, chose organisational transformation as her theme.

She said: “Like sustainability that has a caveat: it is going to outweigh the technology we have been talking about for so many years. Businesses are having to consolidate at this point in time.”

Julie Mitchelmore, VP for Digital, A+E Networks, chose the climate created by the focus on consolidation. She said: “It is almost like safety first. Which is just doubling down on your core strengths.

“There are so many unknowns about this year – the advertising climate, economic climate, and how much customers are going to want to spend on the frivolity of entertainment,” she added.

Krishna Pothula, Head of Architecture, RTL, picked on actionable insights. He said: “There are a lot of companies struggling with that. But it came out as a mood music as well again, so that was interesting to us.”

Generative AI and sustainability: Digging into innovation

Using the term ‘economy and execution’ as shorthand for going back to corporate basics, Harrison wondered if this year would see a break in innovation.

Mitchelmore said: “It does have the potential to be a dull year, but companies that have the ability to introduce the efficiencies but not stifle the creativity will be the ones that come out ahead.”

Pothula added: “We are trying to be cautious, but at the same time we are not putting innovation aside.”

Brown linked her response to the logic of actionable insights. It would not be a year when companies just hold tight.

Read more Artificial intelligence in broadcasting

“You are already starting to see companies using those insights in order to define what the business model will be, and how they use innovation. Those that are successful will be the ones that dig into that innovation,” she said. “That’s going to drive cost savings and it gives that adaptability that we all need.”

Generative AI and sustainability: The biggest headaches

Brooks was asked if there are particular trends in global content consumption. He said: “We have 750 media customers so we get a good perspective. It is innovation with the same caveat as sustainability – it has got to give a return, and be measurable.

“Everybody is trying to save money, so there is a lot of innovation and automation. Actionable insights are a huge part of what people are looking for, because it drives the return,” he added. “There is a lot of talk about data, and that can be overwhelming. We have put a lot of work into advance analytics, actual insights around automatic root cause analysis, which is one of the biggest headaches in our industry.”

With regard to his long experience working with AI Brooks said: “Often times you don’t know what the data is going to tell you. You have a hypothesis, and you get surprised.”

In going upstream Zixi is analysing about nine billion preliminary data points each day. Brooks said: “We have information on almost all the networks out there. We are trying to take that insight and not only provide actual insight on a client’s data, but across all of our clients. We can look for problem areas.”

Mitchelmore added: “Data hugely impacts how successful we can be with our Pay TV platforms. We have been able to get really decent data that helps us shape our propositions, but more transparency would help everyone.”

Pothula added: “The challenge is connecting the data across the supply chain.”

Generative AI and sustainability: Throwing spaghetti at the wall

The impact of consolidation and disruption was passed to Brown. She said: “The data itself is the most important thing, and how do you get the great data lake in order to have all the components? That does hinder the process, and if it is not there maybe some of the consolidation and part of the disruption is happening without all of the tools for making informed decisions.

“It is a Wild West in terms of consolidation. Even with our own data, we are crowd sourcing it in order to make decisions,” she added.

Clumsiness in the creation of commercial models was handled by Michelmore:

“There is always an element of throwing spaghetti at the wall. The industry has these tectonic plates that keep shifting, and while they are moving, it is causing chaos. At the moment it is about a spread bet, but once the movement has stopped things will be stronger, and future proofing for both customers and the business.”

The issue of the great proliferation of content distributors was passed to Brooks. He said: “People need cost effective ways to deliver their content. There is a huge amount of sharing content amongst what would normally be competitive entities, especially in sports. I do think there is going to be content aggregation.”

Will Google and Meta become the data wranglers? “They are so general purpose,” said Brooks. “What we are looking for are very specific kinds of data aggregation, visualisation and insights. That takes very deep knowledge of what we do as an industry.”

Generative AI and sustainability: Hype versus reality

Chat GPT has relegated the Metaverse to number two in the hype charts. How will it impact the creative community?

Brown took on the challenge. She said: “You definitely cannot ignore it. In some ways it might be like FAST, on a bigger scale, but we are all still trying to figure it out. As it is now there is a lot more hype versus reality, but it may be the next big thing.”

Pothula added: “It is a big bullshitting engine, but is very good at that. We should not ignore it because of the popularity it has gained – 100 million users in two months. It is not a breakthrough in technology, but AI made it available for everyone, and that is the breakthrough.”

Brooks said: “It is opening the average person’s eyes to what unsupervised learning can do with AI. But I don’t think it is going to distract us.”

Generative AI and sustainability: All about execution

The last two themes discussed were organisational alignment, and the requirement that sustainability is a bread winner.

Brown tackled the first. She said: “The industry has so many silos, especially in large companies. A lot of times you have one group working on something and you find 18 months later that somebody else in another organisation within your group was working on the exact same thing.

“We have to transform the way that companies work and the way individuals communicate. This starts at the leadership level. It’s all about the execution, and this year is probably when it is going to happen, because the economics are tied to it.” she added.

The sullying of sustainability from responsibility to cash cow was handled by Brooks.

“We talk to a lot of companies and we see how they buy and what they value , and a lot of people want to do the right thing, and focus on sustainability. But if there are no economics around it, they are not going to take a solution,” he said.

“We are very focussed on it. We are getting 5X throughput on compute, which saves a lot of energy, and compute costs. Everything we have now runs on ARM, which takes 50% of the energy that traditional services might take. So, it does give a return,” he added. “Sustainability should be part of all the decision making.”

Read more Webinar: AI and analytics