Deepfakes and AI-generated content call into question the evidentiary value of video, as the importance of physical witnesses grows
The Conversation analysis shows how AI-generated content and deepfakes are calling into question the evidentiary value of video and photography — using the example of the Tesla Roadster and Musk, the author shows why the physical presence of witnesses and demonstrable origin of content are becoming increasingly important.
An analysis published on The Conversation (author Oliver Bray, a researcher in performance and public communication) examines how AI-generated content and deepfakes are undermining the established role of video and photography as evidence that an event actually occurred. As a starting example, the author cites the unveiling of the new Tesla Roadster by Elon Musk, where, in addition to the livestream and recordings from multiple cameras, a live physical audience was also present — according to the author, Musk stated that the audience was present precisely so that no one could claim the footage was AI-generated.
The author refers to Walter Benjamin's concept of authenticity (tied to the presence of the original in time and space) and describes how photography, film, television, and the internet gradually reduced the importance of physical presence, while the recording itself gained its own evidentiary weight. According to the author, AI disrupts this relationship because it can create convincing depictions of events that never happened, and deepfakes can simulate people saying things they never said — the more sophisticated the imaging technology becomes, the less the image itself reveals about its origin.
At the same time, the text points out that physical witnesses are not automatically a guarantee of truth — as an example, it cites illusionist David Copperfield and his 1983 television trick of making the Statue of Liberty disappear, where the audience present authentically confirmed the experience, not the reality of the event. According to the author, the key question is thus shifting from "does it look real?" to a question of provenance — where the content comes from and who vouches for it; a witness can be present and questionable, even though they may be mistaken or lying.
Why it matters
This is a shift in how the authenticity of a video or photograph can be defended in situations where something depends on it — from journalistic verification to legal disputes to corporate presentations. The text suggests that a recording alone may no longer suffice as evidence and that there is a growing need to document the origin of content or have witnesses who can vouch for it; however, the source does not offer specific procedures or tools for verification.
Two audiences, two different impacts
What this means
For individuals
When evaluating the trustworthiness of video or photographic content, it's no longer possible to assume that the recording itself proves the event happened — what matters more is verifiable provenance and who vouches for the content.
For a business
Companies that use video and photographs as evidence (marketing, journalism, legal disputes) need to account for the fact that a recording alone is no longer sufficient as confirmation of authenticity and will need to be supplemented with documented provenance or witnesses.
Risks and compliance More business impacts →Check the original
Event sources
only one source so far · 1 publisher, 1 independent. We count feeds from the same owner only once.