MVA
AEO

SEO Is Dying, AEO Is Unstable, GEO Is Unproven

September 1, 2026 · 5 min read

For years, the search playbook was simple: rank high, get the click, win the customer. If you've been paying attention to your traffic numbers lately, you already know that playbook doesn't work the way it used to.

Two new acronyms have shown up to fill the gap — Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO). Both promise to be the "next SEO." Neither one is there yet. Here's what's actually going on with all three, and what it means if you're building a brand online right now.

SEO is dying

Not dramatically, but steadily. Nearly 68% of Google searches now end without a single click, because AI Overviews and featured snippets just answer the question right there on the results page (SparkToro; Search Engine Land). You can hold the #1 spot for your target keyword and still watch your traffic drop 20–30% year over year, because there's simply less to click on than there used to be.

This is the uncomfortable part for anyone who's spent years building an SEO strategy: the rules didn't change, the game did. Ranking well is still necessary. It's just no longer sufficient.

AEO is unstable

The pitch behind AEO is reasonable — if people are getting answers instead of links, get your content cited inside those answers instead of fighting for a click that isn't coming anyway.

The problem is you can't really tell if it's working. There's no click, no referral link, no session in your analytics to point to. You're optimizing for something you can't directly see. And even when you do get cited, it doesn't last: the sources AI answers pull from can turn over by 40–60% from one month to the next (Yotpo). So you could do everything right in January and be invisible by March, with no clear reason why.

GEO is unproven

GEO takes this a step further — instead of chasing a citation on a results page, you're trying to get mentioned by name inside an actual AI conversation: ChatGPT, Perplexity, Claude, Gemini. The catch is that most of what influences this is content you don't own — a Reddit thread, a YouTube review, a G2 comparison page.

Early research into what these models actually cite shows a clear bias toward brands that are already big and already everywhere (arXiv). If you're a smaller or newer brand, you're not just competing for attention — you're competing against a system that was never built to notice you in the first place.

The bigger issue: AI is breaking things nobody's fully talking about yet

Beyond reshaping how visibility works, AI is introducing problems that hit SEO, AEO, and GEO all at once.

  • Brand hallucinations. AI engines make up things like pricing, features, or return policies about brands with alarming confidence, and because there's no click-through, people just take it at face value (Niara; Am I Cited).
  • Ecosystem flooding. At the same time, the web is filling up with cheap, AI-written content trying to game these same systems — the same pattern already showing up as fake, AI-generated citations flooding scientific literature (Phys.org). More noise makes it harder for AI models, and for actual humans, to tell what's credible.
  • Incentive collapse. If AI keeps answering questions using content it scrapes for free, without sending any traffic back, the incentive to keep publishing good original content starts to disappear. Fewer people writing means less fresh material for these models to learn from in the first place — a loop that quietly starves itself over time.

So what do you actually do

Stop thinking of SEO, AEO, and GEO as three separate jobs. They're one job now, just with more surfaces to cover.

  • What you measure used to be sessions, clicks, and bounce rate. Now it's share of voice in AI answers and citation frequency.
  • Where you show up used to be your own ranked pages. Now it's forums, reviews, video, and third-party mentions.
  • How you write used to be long-form and keyword-heavy. Now it's dense, structured, and fact-checked.

Keep the site technically solid. Write things that are genuinely useful and easy to verify. And pay attention to how your brand shows up everywhere else on the web — Reddit, review sites, YouTube — because that's increasingly where AI models are forming their opinion of you, whether you're in the room or not. Ranking isn't the goal anymore. Being the trusted answer is.