🌟 Editor's Note
For two decades, SEO ruled the courtroom. Rankings were the verdict. Traffic was the sentence. A new defendant has entered the building, and it doesn't crawl the web the way Google does. Generative AI is reshaping how users find answers, how content gets cited, and what it even means to "rank." This case examines the conflict, the evidence, and what marketers have to do to stay relevant in both dockets. Fair warning: some of the evidence is uncomfortable.
Opening Statement: A New Defendant Enter the Courtroom
The court has seen its share of dramatic entrances.
Florida. Panda. Penguin. Each new algorithm update arrived like a new judge taking the bench, rewriting the rules, rearranging the defendants, and handing down verdicts that reshaped entire industries overnight.
This time, the courtroom itself has changed.
On one side stands Traditional SEO, battle-tested, procedural, built on two decades of precedent. It knows the rules. It has survived every update, every penalty, every pivot Google has thrown its way.
On the other side stands something this court has not seen before. Large Language Models (LLMs) powering Google's AI Overviews, ChatGPT's search mode, and Perplexity don't wait for a user to click a link, scroll a SERP, or visit a website. They generate the answer and move on. Whether your content gets credit for that answer is a separate question entirely.
The question before the court: who controls visibility in the age of generative search?
The evidence will now speak.
⚖️ Exhibit A: Traditional SEO
Twenty Years of Precedent Don't Walk Out Without a Fight
Filed: SEO has been the dominant framework for organic search visibility since the early 2000s, shaped by every major Google update from Florida to Helpful Content. Google still processes over 8.5 billion searches per day. Most of those queries still return a traditional SERP alongside an AI Overview.
The argument: Target the right keywords. Earn authoritative backlinks. Optimize on-page signals. Build technical infrastructure Google's crawlers can read, index, and rank without friction. For years, the ten blue links were the only jury that mattered, and brands that mastered this system were rewarded with clicks, traffic, and revenue.
Evidence entered: A Seer Interactive analysis of 3,119 informational queries across 42 organizations found that organic CTR for queries featuring AI Overviews fell 61% between mid-2024 and September 2025, dropping from 1.76% to 0.61%. A separate GrowthSrc study of 200,000 keywords found that position #1 CTR dropped from 28% to 19% in the same period.
Court's finding: That is not a blip. That is a structural shift. The plaintiff is still standing, and its case is not weak. The room just has more parties in it now.
🤖 Exhibit B: AI Overviews and LLMs
A New Kind of Search that Doesn’t Send Traffic the Same Way
Filed: Google officially rolled out AI Overviews to U.S. users in May 2024. By 2025, they were appearing across a significant share of informational and commercial queries. ChatGPT, Perplexity, and Claude go further, functioning as answer engines where users search entirely outside Google's ecosystem. Some of those users will never see a traditional search results page at all.
Charges: Reduced organic click-through rates. Impressions that don't convert to traffic. A page that once ranked #1 for a "how-to" query may now appear as a cited source inside an AI Overview, generating impressions but far fewer visits. There is something genuinely strange about building a resource that gets used without ever getting the click.
Evidence entered: The Seer Interactive study cited above places the informational CTR drop at 61%. On the other side of the ledger, brands cited in AI responses receive 35% more clicks than those that aren't. A brand surfacing inside a generated answer is getting exposure to a user who may never have searched for it directly.
Court's finding: The impact is real. The opportunity is real. Anyone who tells you they have this fully figured out is guessing.
🔍 Exhibit C: GEO Takes the Stand
A New Discipline Enters the Record
Filed: The term Generative Engine Optimization entered the record through a 2023 research paper out of Princeton and Georgia Tech that tested nine content optimization strategies against generative search engines. It is the discipline of optimizing content to appear within AI-generated responses, not just traditional search results.
The argument: GEO runs parallel to SEO. It does not replace it. The foundation is the same one that has always mattered: E-E-A-T — Experience, Expertise, Authoritativeness, and Trustworthiness. What changes is how that authority gets demonstrated. Specificity over generality. Structured answers over buried conclusions. Original data over recycled summaries. Consistent brand mentions across credible sources over any single link or keyword.
Evidence entered: The Princeton paper tested strategies directly and found results: adding quantitative statistics, citing credible sources, and including quotations from authoritative voices each increased AI citation inclusion. Vague, hedged content fared poorly across every model tested.
Court's finding: GEO does not retire the technical SEO playbook. It adds to it. The AI court is strict, and it has high standards for what it will cite.
🏛️ Exhibit D: How LLMs Decide What to Cite
Expert Testimony on the Selection Process
Filed: LLMs are updated through Retrieval-Augmented Generation (RAG) systems that pull live data at query time. Perplexity and ChatGPT with search access cite specific URLs. Google's AI Overviews surface sources from its existing index, filtered for relevance to the generated response. The court has called expert testimony to answer one question: what actually determines which sources make it through that filter?
The argument: Three signals dominate the selection process. Brand authority, measured by mentions and recognition across credible sources, outweighs raw backlink counts. Organic ranking position still acts as the entry requirement for AI citation consideration. And content freshness is weighted more heavily than most practitioners currently account for.
Evidence entered: An Ahrefs study of 75,000 brands found that brand mentions correlate with AI Overview presence at 3:1 over backlinks. Branded anchor text showed a 0.527 correlation with LLM citation likelihood. Domain Rating, the traditional authority metric, was the weaker signal. Separately, 75% of pages cited in AI Overviews rank in the top 12 organic positions. And Perplexity, which is more transparent about its retrieval logic than most, strongly favors content published or updated within the past 90 days.
Court's finding: Backlinks still matter. They are just no longer the lead argument. You have to earn your way into Google's index before the AI models will consider you at all. Traditional SEO is not the alternative to GEO. It is the prerequisite.
⚖️ Final Verdict
The Court Rules on the Future of Search Visibility
After examining the full record, the court declines to declare a winner.
The SEO vs. LLMs conflict is not a zero-sum proceeding. It is a dual jurisdiction, and brands that treat it as either/or will lose ground in both courtrooms.
Traditional SEO is not dead. Authority, trust, content quality, technical integrity: these signals now feed two systems at once. Google's organic index and the AI models synthesizing answers on top of it. That is more work, not less.
The practitioners who win this case will optimize for citation, not just clicks. They will write for comprehension, not just crawlers. They will treat E-E-A-T as the foundation of every content decision, because both courts are scoring on it.
The court has seen this pattern before. Florida. Panda. Penguin. Each rewarded real authority and punished shortcuts. The standard has not changed. The system enforcing it has.
Verdict entered.
The court will reconvene.
