As AI changes how consumers shop and search, affiliate networks are finding new relevance.

That was the message from EMARKETER’s Meet the Analyst webinar, led by senior analyst Max Willens and PartnerCentric CEO and founder Stephanie Harris. They unpacked how AI-driven discovery is challenging attribution models, redefining credibility, and expanding the role of creators within affiliate ecosystems.

“An LLM can’t tell you what CBD oils will make you feel relaxed rather than tired,” said Willens, adding that AI won’t put affiliates out of business, and instead it is “the kind of relationship that will grow symbiotically.”

The value of sorting out attribution

Attribution has always been a friction point in affiliate marketing, said Willens. The channel’s reliance on last-click attribution, a model dismissed by performance marketers, makes it difficult to capture the true value of upper-funnel partners. Now, AI-driven discovery threatens to make attribution even murkier.

“The affiliate marketing ecosystem is really running at AI full speed,” said Willens, noting that 70% of Zenni citations in large language mode (LLMs) stem from affiliate content.

Shopping has gained more share of ChatGPT mobile queries than any other category tracked, above categories like media and entertainment, according to Sensor Tower data cited in an August Bain & Company report.

“[Attribution] is something that eventually is going to get fixed, because there is too much riding on a continued healthful relationship between AI and affiliate for there not to be,” said Willens, “but there are going to be bumps along the way.”

Building credibility in the age of LLMs

Winning over LLMs starts with better attribution and visibility, said Harris. She said marketers need to focus on three things: Discoverability through generative engine optimization (GEO), finding alternatives to last click, and diversifying partnerships and channels.

“Brands, affiliate, and influencer partners all need to think about how they can make their content more visible in a world that is shifting from Google ranking to visibility in the language models,” she said.

With pay-per-click (PPC) and SEO declining, the affiliate model offers stability, but only if brands evolve their content by treating every digital asset as a potential learning source for AI.

“Make sure that the spotlight on your business includes your competitors and you’re put in a favorable light,” Harris said, pointing to FAQs, review pages, and product guides as content that has authority within LLMs.

Refining creator partnerships

While influencers have driven affiliate growth, a trust gap is holding the channel back. Creators who can build credibility and community are central to performance results. Focusing on reach instead of resonance can also lead to higher costs, more fraud, and lower ROI, said Harris.

Some 83% of respondents follow influencers for recommendations, but only 14% trust an influencer more than someone they know in their real life, according to a PartnerCentric survey Harris mentioned.

“There’s a big delta between those two numbers,” she said. “That’s where relatable human connection comes in. Moving the needle doesn’t always mean going to the bigger partner.”

Affiliate publishers are shaping AI-powered recommendations, which points to high growth potential if marketers test new attribution models while building content that pleases LLMs and retains consumer trust.

“The name of the game is going to be creativity and experimentation as ChatGPT and AI chatbots become more commonplace in the consumer shopping journey,” said Willens.