Does authentic footage matter more when AI writes the answers?

As synthetic video gets cheaper and AI summaries answer questions before anyone clicks, two assets are gaining value for any organization with a story to tell. The first is footage of real people that audiences can trust. The second is a clear, factual public record of the work that answer engines can find, read and cite.

Most of our films point cameras at people who are not actors. For years, authenticity in that work was a matter of taste: a style, a grain, a refusal to over-light a face. It is becoming a matter of evidence. That changes what we make, and it changes how we publish what we have made.

Do people trust AI-generated video?

Most people want to know whether they are looking at it, and many doubt they can tell. In a 2025 Pew Research Center survey of 5,023 US adults, 76% said it is extremely or very important to be able to tell whether pictures, video and text were made by AI or by people, while 53% said they are not confident they can. Half said they are more concerned than excited about AI in daily life, up from 37% in 2021.

The worry reaches beyond the US. The 2026 Edelman Trust Barometer, fielded in 28 countries, found that 37% of respondents cite the growing use of generative AI platforms as an event affecting their trust. When an audience has learned to doubt what it sees, a beautiful image no longer earns belief on its own. It has to be traceable to someone.

Why does documentary footage matter more in the AI era?

Because a real person in a real place, speaking in their own words, comes with a chain of accountability that a generated image lacks. There is a name, a location, a date, a crew, and a person who can say "that was me."

That thread runs through the work we are proudest of. For LIVING, made with Bristol Myers Squibb, we filmed patients in rural China, Sweden, France and the United States in a cinéma vérité style, and set aside the staged, sanitized imagery the pharmaceutical category tends to use. The product stayed out of the frame. The film went on to win the Grand Prize for Best Overall Video at Ragan's 2023 Video, Visual & Virtual Awards.

For The Exonerated Five, a film made with the Harris-Walz campaign in 2024, Korey Wise, Yusef Salaam, Kevin Richardson, Antron McCray and Raymond Santana spoke directly to camera. We used no music and no graphics. In Zambia, our observational film for the World Bank's Partnership for Economic Inclusion follows Fitness, a mother, entrepreneur and savings-group member, and her daughter Venus, through ordinary days.

When anyone can generate a face, the face that can be traced back to a life becomes the scarce thing.

The craft lesson is subtraction. Every layer of polish added in the edit is a layer an audience now has to see through. The more a film looks like it could have been prompted, the harder it has to work to be believed.

How do AI Overviews affect website traffic?

The early evidence says they reduce clicks, though the measurement is disputed. A 2025 Pew Research Center study of the browsing data of 900 US adults, reported by The Register, found that users clicked a traditional search result in 8% of visits when an AI Overview appeared, against 15% when none did. Only about 1% clicked a source link inside the summary itself. Google disputed the study, calling its methodology flawed.

Chatbots are becoming a route to information too. The Reuters Institute Digital News Report 2026 found that 10% of people use AI chatbots for news each week, up from 7% the year before; among under-35s the figure is 16%. Only 20% say they trust chatbot answers. In 2024, Gartner forecast that search engine volume would fall 25% by 2026. That was a prediction, not a measurement, and it has been challenged. The direction is clear enough. For a growing share of questions, a summary is the front door, and your page may be read by a machine before a person ever sees it.

How do organizations get cited by AI search?

Our working view is that the best preparation is a public record that is specific, consistent and checkable, with each question set beside its answer. An answer engine can only repeat what it can find. If a case study gives two different view counts for the same film, or a date that disagrees with the event's own records, the summary will inherit the confusion.

We are applying a few rules to how we write about our own work:

• Say who made what, for whom, and when. Credits, partners, places and dates belong in plain sentences.
• Label the source of every outcome. An award confirmed by the awarding body reads differently from a figure in a client's own report, and both read differently from our own count.
• Answer the question people actually ask. A heading phrased as a real search, followed by a direct answer in its first sentence, is easier for a reader and a machine to use.
• Correct the record. Our own archive is still being cleaned up on this front, and we would rather fix a date than have it repeated with confidence.

None of this replaces the work itself. A clean record of a forgettable film is still a forgettable film. But the two disciplines turn out to be one: show real people truthfully, then describe what you did truthfully. You can see how that plays out across our work and in our original documentaries at HUMAN__stories™.

Questions people ask

What is answer engine optimization?

Answer engine optimization is the practice of making content easy for AI systems that answer questions directly, such as search summaries and chatbots, to find, understand and cite. In practice it means clear questions, direct answers, consistent facts and visible sources. It rewards organizations whose public record is accurate.

Can audiences tell AI-made video from real footage?

Many people doubt they can. In Pew Research Center's 2025 survey, 76% of US adults said being able to tell AI-made content from human-made content is extremely or very important, but 53% were not confident they could do it. That gap is why traceable, documentary footage of real people carries more weight than it used to.

What should an organization publish so AI search describes it accurately?

Publish specific, consistent facts about your work: who was involved, where, when and what happened, with outcomes labeled by source. Phrase key pages around the questions people ask and answer them in the first sentence. Fix conflicting figures, because a summary will repeat whichever one it finds.

If your organization needs stories built on real people and a record that holds up, apply to work with HUMAN__agency™.