As generative tools make it progressively cheaper to produce synthetic content and synthetic spokespeople, the obvious expectation is that advertising will migrate in that direction. Lower production costs, faster turnaround, infinite variation, and no scheduling constraints represent exactly the efficiencies marketing departments are usually rewarded for capturing.
A meaningful number of brands appear to be betting on the opposite and the reasoning behind that bet is worth examining because it is not obvious
Street Poller Media Founder and CEO Shane Ginsberg has put the underlying tension bluntly, “you cannot scale your way to trust.” His argument is that the qualities audiences respond to in advertising, credibility and the sense that something genuinely happened, are precisely the qualities that do not survive mass production. This means the efficiencies generative tools offer come at the direct expense of the outcome the advertising was meant to produce.
As an illustrative example, Ginsberg has pointed to brands like Rolex, which does not run AI-generated advertising on Instagram, instead sponsoring tennis and golf events including the PGA Tour, because that is where its audience actually spends attention. This maintains a social presence built around real moments rather than synthetic content. While Rolex is not a client relationship, Ginsberg uses it purely as an illustration of a broader instinct playing out across advertising. Even as companies race to automate content production, the brands with the most accumulated equity to protect are choosing formats that read as unmistakably human.
Ginsberg has also drawn a direct line to consumer fatigue with AI content specifically. In his assessment, audiences are growing tired of AI-generated marketing and automated content. This is notable because the reason is not that the output is poorly made. In fact, quality has improved rapidly enough that production value is no longer the tell. The fatigue stems from the fact that synthetic content does not connect the way an authentic unscripted conversation does. The distinction has an important implication; the problem may not be solvable through better production.
This dynamic points toward a possible bifurcation in advertising rather than wholesale replacement. Content where authenticity is not the operative mechanism, product demonstrations, explainers, animation, and anything primarily informational, seem likely to automate substantially. Content where the entire value depends on a viewer believing a real person had a real reaction may become more valuable precisely because that quality becomes scarcer as everything around it becomes synthesizable.
The available platform data offers some support for the direction, if not for the specific prediction. Content that reads as organic rather than overly-produced continues to significantly outperform traditional advertising formats across short-form video platforms. Algorithm changes at Meta and elsewhere have also increasingly rewarded creative that is not overly sales-centric.
It is worth noting that audiences have adapted to every previous shift in advertising aesthetics and a format that reads as authentic today may read as a recognizable convention within a few years. At which point its advantage erodes on its own terms rather than because of anything AI does.
As synthetic content becomes cheap and ubiquitous, verifiable evidence that something actually occurred becomes a scarcer commodity, Scarce commodities tend to appreciate.
