• by Laurie Sullivan  @lauriesullivan, December 27, 2024

Stagwell's media agency Assembly predicts that in 2025 more advertisers will use a strategy called anti-algo, a concept related to approaches that are the opposite of what is expected by introducing breaks in algorithmic reasoning,

As AI integration becomes more pervasive, the appeal of "anti-algo," or anti-algorithm, will surge as a result of AI, both generative and general.

Assembly's prediction suggests consumers will seek organic, unfiltered interactions with culture and content, looking for authenticity and spontaneity rather than algorithm-driven experiences.

The shift will challenge brands to find new ways to connect with audiences in more genuine ways.

The concept of anti-algo is not new. It has been used by some for years in a variety of ways, but not so much in advertising. Assembly predicts more advertisers and marketers will adopt the practice next year.

“We're slated to test one this concept in 2025, for a QSR brand,” Jonathan Kagan, director of search and digital media strategy at Amsive, told MediaPost. “It is still a rather unknown and not highly used, as it goes against the concept of in-market and qualified audience targeting that advertisers heavily rely on.”

Kagan first heard the term in 2019 while on the Reddit platform. The first time he heard the term related to advertising was in 2021 from an MIT grad student, Anna Woorim Chung.

“She wrote about it in a paper published about subverting algorithms on social media,” he says.

It is a bit of a stretch, he said, questioning how big of an adoption the industry will see, but there are some very specific and niche use cases where it can be applied, Kagan says.

Kagan has seen a couple of use cases, including one that goes back to 2012 before the concept even existed. He now understands how it could be used for reaching new and prospective customers, who do not fit the norms of a current in-market and affinity-based behavior audience.

The anti-algo strategy centers on when someone shows a propensity of liking specific content, but rather than serving them that content, the marketers will instead serve them ads that are for something else, which does seem a bit strange.

For example, if someone searches for a restaurant, an anti-algo strategy would highlight the views of restaurant workers instead of food critics.

Jim Yu, CEO at BrightEdge, described how anti-algorithms can play an important role in generative AI search systems by enabling the engine to deliver well-rounded responses that reflect diverse perspectives. While algorithms offer numerous benefits, they can also reinforce self-perpetuating conclusions.

“Over time, this can gradually narrow the range of search results, increasing the likelihood of specific decisions being favored, even if those aren't necessarily optimal,” he said. “To mitigate this bias, breaking information chains, considering unconventional or outlier data, and integrating anti-algorithms can help present alternative viewpoints and achieve a more balanced perspective overall.” 

For example, when looking for ways to navigate New York City, an algorithm might prioritize mapping apps and user-generated content from tourists. In contrast, an anti-algo approach might connect you with locals, seasoned tour guides, or taxi drivers who intimately understand the city's layout.

“AI search engines, however, must exercise caution when implementing anti-algorithms,” he said. “In some cases, introducing alternative viewpoints can pose risks — such as in scenarios involving sensitive topics like vaccines, where promoting fringe perspectives might be harmful.”

Not all agency experts believe anti-algos will become a thing in 2025. Some think it makes little sense to use it as a targeting technique.

"I suppose if we had data that said a user hates the other one, we could consider running an ad for the hated product right before we ran the ad for the loved product,” said Marty Weintraub, founder at Aimclear. “Still seems like kind of a reach.”