Pat Maddox on AI and Video Production in Weather Reporting

Artificial Intelligence (AI) is transforming video production for weather reporting. Pat Maddox discusses AI’s role as a complementary tool in content creation, exploring ethical considerations, tailoring content for different platforms, and predictions for more personalized weather experiences. Maddox’s insights offer a compelling glimpse into the future of weather media, where technology and human expertise combine to deliver more engaging, tailored experiences for viewers.

Q: Tell us about yourself. What has your career trajectory been?

A:  I’m one of the video leads at The Weather Company, and my day-to-day is overseeing all video production in terms of performance, creation, ideation, and execution. I started my career in long-form feature films but realized a few years into that path that I preferred working on smaller productions. It takes years to get involved in creative decision-making in traditional long-form productions, but in small projects, I wore multiple hats. Then, that coincided with the release of the DSLR movement, cameras like the 5D Mark II came out, and video content started appearing on all social media channels. I was in the right place at the right time with a few other people, and we started doing behind-the-scenes content and fashion-style content for Instagram and other social platforms. That allowed me to take on larger, more creative projects and execute from the brief all the way to distribution.

Q: In what ways has artificial intelligence changed the landscape of video production? 

A: Any element of video production now has AI developments. It’s not always said in those words, but Premiere Pro—one of our leading editing software—has it integrated into their captions. At The Weather Company, we don’t use it to generate visuals. I think all filmmakers and creatives probably use it for ideation more than anything.

It’s a good ‘sounding board’ rather than a tool you can rely on to execute elements of video production. You cannot entirely depend on AI at this point for any position or in any phase of video production. It’s just another tool that people are using as a reference point.

Q: What benefits and challenges have you encountered in integrating AI into the daily production processes?

A: One of the benefits is that you can use it as a reference tool to make sure that things are spelled correctly, for grammar, as a thesaurus, or as things that people have always used Google for. But one thing that I don’t think AI can do is replace experience. In video production, experience is crucial in terms of what you’ve done and what you can forecast for the industry’s future.

Those of us who’ve worked in video production for a while have seen so many changes that we know 10 or 20 years from now, things will be completely different again. That’s just the nature of the job. It’s hard to forecast precisely where we’ll see the most significant impacts. However, it’s easy to say that there won’t be any elements of production that won’t have some sort of collaboration with AI.

Q: What new skills have you had to develop to keep pace with AI and technology advancements in your field?

A: It comes down to the user’s experience–how well can they write a prompt and how well they can interact with AI? What you put into it is what you get out of it. So, the better context you can offer AI, the more you will get in return. It is more about learning how to use the tool correctly, knowing it’s in your tool belt, and accepting that it’s here and being used. If you’re not going to use it, other people will. So, it needs to be something that you understand.

Q: What future advancements in AI are you most excited about for weather reporting? How do you see AI evolving in your operations over the next decade?

A:  It will vary from place to place, but we will see more avatars. In some industries within the world of content creation, we’re seeing a lot of avatar generation now that appears ‘reality-based.’ With things like the Apple Vision Pro coming out and more wearable tech, it’s only a matter of time before augmented reality is involved in everything we do. And in that world, AI will have a huge say in how things are done.

Going forward in the next ten years, I would probably guess that we will also see more personalization in how AI communicates with individuals. I’m sure we’ll see applications where AI can understand the cadence and communications of people and then use that to communicate with them directly. When the Apple Vision Pro came out, we explored how that would impact forecasts ten years from now and what that would look like. I have no idea right now, but it’s one of those things that makes me very curious about how that world changes. Wearable watches are one thing, but with goggles, we will go in directions that are pretty hard to predict.

Q: What ethical considerations do you think through when deploying AI for video production?

A:  So much! AI is a powerful tool, but you have to be careful. We always see deep fakes on Twitter, and in the news, not too long ago, there were stories about Scarlett Johansson and ChatGPT. It’s such dangerous territory that one of the big things everyone has to consider when they get into using AI is its purpose. From a generation point of view, our team at The Weather Company is known for our forecast accuracy, so we can’t generate a visual and say, ‘This is a wildfire.’ We need to show the actual wildfire. You must share truths and facts when it gets into the news media world.

In content, there have always been ethical considerations: who you’re going to film, how you’re going to film them, etc. We just need to keep applying that same care, consideration, and respect to the process that we’ve always had. Just because AI is involved doesn’t mean you should do things quicker. If anything, it means you should do things slower because if you try to remove opinions and people from process elements, you eliminate experience.

Q: What factors do you consider when deciding the content of a video to ensure it is engaging and impactful? 

A: Great question. These days, everything comes down to distribution and audience. Once you know where you’re distributing and who your audience is, you can backtrack to it. We know very well from our data dives that the way we serve our TikTok, Facebook, and app audience is entirely different. For example, we traditionally see weather and animal content perform well on Facebook, so we use that medium for those types of content.

Q: How do you balance creativity with the need for accurate and safety-focused information in weather-related content?

A: When distributing content via The Weather Company app, we know that safety is the priority. We ensure that we are accurate and that everything we say is precisely what we want our users to know. We’re missing the point if our users aren’t gaining safety information. We’re not ever putting creativity aside, but it comes in different forms and approaches depending on who we’re talking to.

Q: How do you keep up with new video production technologies and techniques?

A: It’s always been fundamental to the world of content creation to know what new tech is coming. Before AI, there was everything else: from camera operators to robotic cameras to digital integration. It’s always been an element that emerging producers and content creators have had to be cognitive of because if you weren’t aware of the new tech coming, there was a good chance someone younger than you or someone emerging into the industry would know it.

I think it’s essential to look at content creators on YouTube, watch high-quality feature films, and consume content on every platform. It’s the best way to learn what’s happening in the field. As much as I don’t have a personal TikTok account, I still have to keep up because that’s where emerging media is. It’s important to always see what new styles are being used.

It’s just as crucial, too, for a content creator to watch something like Oppenheimer, last year’s highest technically achieving film. Same with Dune and Dune: Part Two. You need to have an understanding of all production sizes. Having a scale and understanding of how you can do something at the cheapest level and how it’s done at the most expensive level helps when you’re creating content.

Q: What advice would you give to a forward-looking brand looking to level up its video production processes utilizing AI?

A: AI is just an additional tool that will be added to the creative’s tool belt. We’ve always seen new tools added and, ultimately, old ones removed from that tool belt. So, I don’t think you can say to a new company that you can lean on AI to do everything. Fundamentally, you’ll need someone with quality and experience to control the AI to get anything out of it. If you’re going to rely on AI alone, you’re probably going to end up unhappy with your output. But if you have someone controlling it, you’re confident and know what your brand wants and its purpose. You can’t build an entire content department with the idea of just using AI. It’s all about the people who are controlling it.

Q: What is the most salient advice you’ve received in your career? What advice would you offer new professionals entering the field with an interest in AI and video production?

A: Get out of the way of good content. It doesn’t matter what field you’re in or how the content is being created. So many times, we want to put a creative spin on things when all the pieces are already there for us. People want authenticity in everything they’re looking at. There’s no point putting bells and whistles on something unique because all you’re doing at that point is distracting from what people want to see—great content. Just let it do its purpose. People at this point in the world have so many choices, especially regarding digital media.

In terms of advice to new people entering the field, know your audience. What tools you use doesn’t matter if you’re not communicating to your audience and don’t know your distribution purposes. There’s no value in using the best camera in the world if you’re just making content for a small vertical audience on a phone. You can do that with a phone or a high-quality, small camera.

If your audience is on a giant movie screen, you may need to scale up, but it’s all about knowing what you’ll do with your content. Once you know what you will do with it, you can fill in the gaps of how you will create it. But know your destination before you set off on your journey, and there will be a million different fun stops in the middle, as well as things you’ll learn on the way. But you need to know your deliverables first.