Different AI platforms, different sources: How can brands stay ahead in the citation game?

Consumers are increasingly turning to AI answer engines to discover and compare brands, but the platforms shaping those recommendations are not all reading from the same playbook.

A recent report by VaynerX and Profound found that while social, creator and user-generated content are becoming a growing source of citations across major AI platforms, each answer engine builds authority differently. Google's AI products lean heavily on YouTube, ChatGPT draws more from Reddit and review platforms, while Microsoft Copilot increasingly incorporates signals from LinkedIn and Microsoft's broader ecosystem.

The study, which analysed thousands of brand recommendations across six AI answer engines, also found that AI systems are learning at an unprecedented pace.

The median time between publication and citation is just 6.8 days, with 90% of content cited within approximately 37 days.

The findings suggest marketers may need to move beyond a one-size-fits-all approach to AI optimisation as discovery becomes increasingly fragmented.

So, if different AI platforms are building trust in different ways, how should marketers adapt their strategies to stay discoverable across an increasingly fragmented AI landscape?

Don’t miss: Showing up in AI answers isn't enough if audiences don't believe them

For the marketing and advertising industry, the report signals more than just another shift in search behaviour. According to Jacky Chan, CTO of Votee AI and Beever AI, brands are moving from search engine optimisation (SEO) into what he describes as generative engine optimisation (GEO), where success depends not on satisfying a single search algorithm, but multiple AI models that each consume information differently. He added: 

The old game was pleasing one algorithm. The new game is feeding many AIs, and each one eats from a different plate.

Rather than asking how to rank on Google, marketers should now be asking what each AI model reads, and whether their brand is present within those ecosystems, Chan added. With AI systems citing fresh content in a median of under seven days, he argued this is no longer a campaign-by-campaign exercise but an always-on discipline.

Not everyone, however, sees this as an entirely new playbook. Dominique Rose Van-Winther, chief AI evangelist and CEO of Final Upgrade AI, believes much of AI optimisation still rests on familiar search fundamentals.

She noted that while different AI models may lean on different ecosystems today, those relationships continue to evolve as platforms change how data is accessed, scraped and indexed. 

"The underlying point still holds where if you've spent a decade optimising only for Google, you can be missing from a big share of the answers—not because the AI dislikes you, but because you were never in the index it was reading from," she added. 

Whether marketers call it SEO or GEO, Pamela Phua, managing partner, Southeast Asia at Media OutReach Newswire, believes the underlying shift is clear, "The fundamental game has changed and we are no longer optimising for search engines, but for answer engines."

To succeed, brands need to move beyond human search intent and ensure their content is structured in ways AI systems can easily interpret, contextualise and accurately cite, with machine-readable formats and structured metadata becoming just as important as compelling storytelling.

The new challenge is trust, not visibility 

Appearing in AI-generated answers is only half the battle. As different answer engines increasingly rely on different sources, experts argue the bigger challenge is ensuring those systems arrive at a consistent understanding of a brand.

"The deepest challenge is that your reader is no longer human," Chan said. "Your website used to persuade a person; now it has to persuade an AI that's doing the research for that person."

Unlike traditional search, AI does not present users with multiple links to compare. Instead, it synthesises information from across the web into a single response.

That means brands have less control over the final recommendation and more responsibility for ensuring the information AI consumes is accurate, consistent and trustworthy. For Chan, that extends well beyond owned media.

Consistency now depends on whether creators, influencers, review platforms and third-party publications are telling the same story because those are increasingly the sources AI trusts.

Van-Winther describes the resulting challenge as "narrative drift". Different AI engines, she explained, can develop entirely different understandings of the same brand depending on the sources they access.

One engine may primarily associate a company with YouTube tutorials, another with Reddit discussions, while a third draws mostly from LinkedIn or analyst coverage. She noted: 

Same brand, three reputations.

Rather than trying to force identical outputs across every AI platform, Van-Winther believes marketers should focus on creating a strong, consistent source of truth that can withstand those differences.

Building authority in the AI era

If consistency is becoming the foundation of AI trust, the next question is how brands can earn that trust in the first place. While publishing frequently remains important, the experts agreed that simply producing more content is unlikely to improve AI recommendations.

Chan warned that flooding the internet with content has become increasingly ineffective, and potentially dangerous, as fabricated information and AI manipulation become easier to produce.

Instead, he argued that brands need to understand how AI retrieves, filters and ranks information, ensuring they appear where answer engines actually look for evidence rather than simply generating more material.

Phua echoed that sentiment, arguing that authenticity and technical architecture will increasingly separate brands that consistently appear in AI recommendations from those that do not.

"While maintaining content volume remains important to stay relevant, volume alone is no longer a differentiator—authenticity and technical architecture are," she said.

According to Phua, the brands that consistently win AI recommendations will be those that master AI trust signals, with structured metadata, authoritative news distribution and earned media across high-authority publications all reinforcing credibility in the eyes of AI systems.

For Van-Winther, the focus should be less on gaming the system and more on building lasting credibility.

"Everyone wants the hack for the new system," she said. "But the ones who win are doing the thing nobody wants to hear: Being the best answer, provably."

Rather than relying on marketing claims alone, she argued AI increasingly rewards independent validation, proprietary data, expert attribution and consistently corroborated facts. When information is well structured and supported across trusted sources, it becomes easier for answer engines to retrieve, verify and recommend.

Ultimately, while the platforms consumers use may continue to evolve, the experts agreed that the underlying objective remains remarkably consistent. Marketers may no longer control the answers AI gives consumers, but they can still influence the evidence those answers are built on.

As Van-Winther put it:

You're not controlling the output anymore. You're controlling the inputs the system builds it from.

Be part of Digital Marketing Asia Singapore on 22–23 September 2026 to discover the strategies, technologies, and real-world lessons helping brands scale AI and orchestrate predictive customer journeys.

Related articles: 
What's behind the disconnect splitting the AI narrative?   
Are AI chatbots building the next walled garden?  
Brands struggle with AI disclosure as usage surges across marketing