This is how we created our first corporate video with generative AI

We have long been hearing that productivity gains are one of the main advantages offered by Artificial Intelligence. However, in day to day operations, productivity can feel somewhat abstract and it is not always easy to clearly see that value.

For this reason, we wanted to take this opportunity to share a real case, told in the first person, where we can clearly demonstrate the benefits we so often advocate for those of us who work with this technology.

A corporate video with one condition: it had to be different

A few months ago, our communications department offered us the opportunity to produce a promotional video for our business unit, with the aim of showcasing our capabilities and experience to the world.

This is a typical initiative in a large corporation. Companies like Telefónica Tech frequently use this format to share all kinds of content with our clients, users and the media.

Up to this point, everything was business as usual. However, this time we wanted to do something different and, above all, something consistent with our principles and our value proposition.

When you advocate AI, you have to apply it

As I mentioned at the beginning of this article, within the AI & Data area at Telefónica Tech we dedicate our time to helping clients unlock the full potential of Artificial Intelligence across multiple domains. For this reason, now that we had the opportunity to create new content, it would have been difficult to explain if we had missed the chance to use AI as the main tool.

We wanted to do something different, but above all something consistent with who we are and what we offer our clients.

No sooner said than done. The decision was made. The new video for our business unit would be generated with Artificial Intelligence. But once the decision was taken, the real challenge began. Doing something for the first time, no matter how exciting it may be, comes with difficulties that must be addressed to achieve a successful outcome.

In our case, we started working along two parallel lines. The first was to decide what we wanted to say and how we wanted to say it. A business area with more than 10 years of history, like ours, with thousands of projects behind it, has a lot to share and not everything fits into a 90 second video. The second was to choose the right partner for the job.

From message to partner: two key decisions

The first task was straightforward. We are fortunate to have a communications team that understands our work perfectly and knows our target audience just as well. The second was not necessarily complicated, but it did generate greater uncertainty.

No one in the company had yet worked with a specialized agency focused on video creation using generative Artificial Intelligence. We had to start from scratch and search for providers ourselves. And this is where we encountered the first obstacle.

Generative AI in video: still an uncharted territory

Today, there are still few agencies truly specialized in the use of generative AI for video creation. And I say specialized because most content agencies are moving in that direction. However, for a large corporate video, it is essential that the provider has full command of the tools. The quality must be outstanding.

Doing something for the first time with a new technology means embracing uncertainty, even when its potential is clear.

This lack of experience, which is understandable given how new the technology is, means there is no established industry standard. There is no single “right” way of doing things, nor is there even a standard way of pricing a project of this kind.

We reviewed multiple providers and, to keep it brief, we can share that we ultimately decided to work with MITO Films. We made the decision with some hesitation, given the uncertainty of working with a new provider in a field we had never explored before. But now that everything is complete, there is no doubt it was the right choice.

But let us return to productivity.

Before AI: speed or customization, but not both

Before the arrival of generative AI, there were essentially two ways to create a corporate video: use stock footage or produce your own video.

The first option was fast and affordable. Stock libraries offer thousands of ready to use assets at a reasonable price. However, in return, they provide no real customization options, making it extremely difficult to properly convey a brand’s values.

The second option is slower and more expensive. Producing your own video with actors and real locations requires a significantly greater investment in terms of time, resources and budget. That said, the result is exactly what you expect. A tailor made piece that, if done well, faithfully reflects the ideas and values you want to communicate.

Until now, creating a corporate video meant choosing between speed and customization. With generative AI, that trade off is beginning to disappear.

As I was saying, until recently these were the only two possibilities, and you had to choose one, giving up the advantages offered by the other. We can confirm that this is no longer the case.

I will pause briefly here to share the link to our video (in Spanish). I believe that after watching it, some of the ideas I will discuss next and that we have experienced first hand will be much clearer.

Advantages of creating a corporate video with generative AI

I would like to focus mainly on the advantages of using generative AI that we have observed compared to past experiences:

• Customization: The level of customization is exceptional. We can adjust every detail in every frame. Spaces, colors, lighting, logos… This is impossible with stock footage and very costly with recorded videos.
• Flexibility: The ability to make changes is enormous. With a simple prompt, we can replace a man with a woman in just minutes. Or change the background from an airport to an office. With recorded videos, this would require hours of editing or even reshooting the scene with new actors and new locations.
• Speed: Obtaining the first version of the content is almost as fast as using stock footage. Recorded videos require planning, filming and editing. It takes much longer before you can see an initial version.
• Error tolerance: AI allows you to experiment and make mistakes with almost no consequences. Discarding an image has virtually no cost, which provides significant peace of mind. In a recorded video, sometimes you cannot discard content simply because there is no alternative. With AI, however, we can generate as many new clips as we need and choose the one that best fits.
• Cost: Without sharing specific figures, I can confirm that in this case cost has not been the main advantage. However, we have no doubt that as this technology becomes the industry standard, costs will continue to decrease.

And here is the final video (in Spanish).

Conclusion

Having gone through this experience, at Telefónica Tech we are convinced that the emergence of generative AI in the field of video creation represents a true revolution. It combines the best aspects of customization and quality from recorded video with the execution speed and flexibility of stock libraries.

In terms of productivity, there is no doubt that generative AI is the winning option.