Organic traffic declines. The team panics. Someone says the words “generative engine optimization.” A new initiative spins up: get the brand cited by ChatGPT, show up in AI Mode, win in Perplexity. The tactics are familiar if you’ve been paying attention. Self-promotional listicles. Hidden prompt injections. Content engineered not for readers but for retrieval.
It works, briefly. Then Google notices. Rankings drop further. And here’s the part nobody’s talking about: when your Google rankings drop, your AI citations drop too.
New data from Lily Ray makes this painfully concrete. She tracked eleven company blogs that got hit by Google’s unconfirmed January 2026 algorithm update and pulled their AI search citation data across four platforms. Every single site that lost organic visibility also lost AI citations. The average decline was 22.5%. The panic move to “optimize for AI” didn’t just fail to compensate. It may have accelerated the damage.
ChatGPT Follows Google More Than Google’s Own AI Does
The most striking finding isn’t about Google’s products. It’s about ChatGPT.
ChatGPT citations dropped 27.8% on average across the eleven sites, with some subfolders losing more than 40%. That’s a steeper decline than Gemini, which is an actual Google product drawing from Google’s own index.
The implication is hard to miss. ChatGPT’s web retrieval leans heavily on Google’s search results. When Google demotes a page, ChatGPT stops citing it too, often more aggressively than Google’s own AI tools do.
This is exactly what makes the cycle so vicious. The platform everyone is scrambling to optimize for is the one most tightly coupled to the rankings they’re destroying in the process.
The Compounding Loop
Here’s how it plays out in practice. A B2B SaaS company sees organic traffic declining. Leadership wants answers. The content team pivots toward “AI visibility,” publishing listicle-style content designed to get scraped and cited by LLMs. Google’s algorithms flag the content as low-quality or self-promotional. Organic rankings drop further. Because ChatGPT’s retrieval pipeline runs through Google’s index, the AI citations disappear too.
Now the company has lost ground on both fronts. The original organic traffic problem is worse. The AI visibility they were chasing never materializes in a durable way. And the content that caused the damage is still sitting on the blog, continuing to drag down domain quality signals.
Ray’s data shows this isn’t theoretical. The sites in her study saw organic traffic drops of 6% to 53%. ChatGPT citation losses tracked closely, sometimes exceeding the organic decline itself. One subfolder lost 42.3% of its ChatGPT citations.
Perplexity Is the Exception That Proves the Rule
Perplexity behaved differently. Only four of the eleven sites saw citation declines there. Seven actually saw citation growth during the same period. This lines up with the industry’s working theory: Perplexity pulls primarily from non-Google sources, likely the Brave Search API combined with its own crawler (if you are a paid subscriber to the information, you can read more about Brave here too).
That’s interesting, but the scale difference matters. ChatGPT pulled 5.8 billion web visits in August 2025. Perplexity pulled 148 million. The one platform that’s independent of Google is roughly 39x smaller. For most companies, the overwhelming majority of AI-driven visibility still flows through a pipeline where Google’s rankings dictate the outcome.
The Uncomfortable Takeaway
The “GEO” movement has been selling a comforting story: that AI search is a new frontier, separate from traditional SEO, requiring its own playbook. Ray’s data suggests something less convenient. Google’s organic index isn’t just powering Google search anymore. It’s the backbone of most AI-generated answers across platforms. When Google decides your content doesn’t deserve to rank, that judgment cascades through ChatGPT, AI Mode, and Gemini almost simultaneously.
The practical implication is that any tactic which risks your organic rankings isn’t just an SEO gamble. It’s an AI visibility gamble too. And the teams most aggressively chasing AI citations with short-term tactics are the ones most likely to trigger the exact cycle they’re trying to escape.
The priority order hasn’t changed. Build content that earns real organic visibility. The AI citations follow the same signal. There is no shortcut that doesn’t eventually cut both ways.