A year ago we were still debating whether AI would change search.
That debate is over.
Google's AI Overviews now appear on roughly 30% of searches. ChatGPT handles billions of queries monthly. Perplexity doubled its user base in 2025. The way people find information shifted fundamentally.
The old playbook doesn't work. You're not just competing for ten blue links anymore. You're competing to be the source an AI chooses to cite.
What AI search looks like right now
Let me get specific about what changed.
Google AI Overviews show up on about 25-30% of US searches now, up from around 11% in early 2025. For informational queries, it's higher. For commercial queries, Google stays more conservative. But the trend is clear.
ChatGPT is still the dominant AI assistant. Search is baked into the product. Someone asks a question, it pulls from the web, synthesizes, and delivers an answer. Sometimes with citations, sometimes not.
Perplexity carved out a niche as the "research" AI. Cleaner source attribution than competitors. For product research, technical questions, current events, it's becoming a first choice.
Zero-click searches are accelerating. When AI answers directly in the interface, people don't need to click through. This isn't speculation. I can see it in my traffic data.
But here's what the doomers get wrong: citation still matters. When AI cites your content as a source, you gain something potentially more valuable than a click. Trust. The AI is vouching for you.
What actually changed (and what didn't)
Some things shifted. Some stayed the same. Here's what I've noticed matters.
Authority signals are non-negotiable now
In 2024, you could sometimes rank with thin content and minimal backlinks if you hit the right keywords. Those days ended.
AI systems lean hard on source authority when deciding what to cite. They're pattern-matching against:
- Which domains get cited by other credible sources
- Who Wikipedia references
- What trusted publications link to
- Which content shows up in top organic results
The research backs this up. Over 99% of sources cited in Google's AI Overviews come from pages already ranking in the top 10. AI isn't discovering hidden gems. It's amplifying already-authoritative sources.
Building quality backlinks isn't optional. It's the price of admission.
E-E-A-T is everywhere
Google's E-E-A-T framework became the lens through which AI evaluates content.
Not just what you write. Who writes it, what credentials they have, whether the rest of the web treats you as credible.
Practical stuff:
- Visible authorship matters. Anonymous content struggles.
- First-hand experience differentiates. AI can't fake having used a product.
- External validation compounds. Links from trusted sources signal authority.
If your content strategy is "publish and pray," you're already behind.
Quality bars went up
Google's March 2024 update specifically targeted "scaled content abuse." Mass-produced, low-quality content regardless of whether AI or humans created it. The trend continued through 2025.
Thin content that might have ranked in 2023 now gets buried. AI-generated slop without human oversight gets filtered.
What wins: original research, first-hand experience, clear answers to specific questions, comprehensive coverage from credible authors.
What loses: regurgitated information, keyword-stuffed pages, content that reads like it was generated in 30 seconds.
AI tools worth using (and which to skip)
Every SEO tool added "AI features" in 2025. Most are gimmicks. Some are genuinely useful.
For content optimization
Clearscope, Surfer, MarketMuse analyze top-ranking content and suggest how to optimize yours. Helpful for topical coverage. But following their suggestions blindly produces formulaic content that all looks the same.
Use them for finding gaps. Ignore them when they suggest adding information outside your expertise.
For research
Perplexity is genuinely useful. It aggregates sources faster than manual searching. Just verify everything.
ChatGPT with search works for brainstorming. Fact-check aggressively though. It hallucinates less than before. Still hallucinates.
For content generation
Here's the uncomfortable truth: AI-generated content can work. It can also destroy your site.
The difference is quality control. AI as a starting point, heavily edited by humans with expertise? Viable. AI as a content factory with no oversight? You get filtered.
Google doesn't care if content is AI-generated. They care if it's helpful. The March 2024 update made that distinction painfully clear.
What's overhyped
- "AI SEO agents" claiming to automate your entire strategy
- Tools promising to "guarantee" AI citations
- Automated content spinners with fancy new branding
- Anything claiming proprietary access to Google's algorithm
If it sounds too good to be true, it is.
How to get cited by AI
This is the game now. Not just ranking, but becoming a source AI systems trust and cite.
Be directly quotable
AI systems extract and quote specific statements. Vague, meandering content doesn't get cited. Clear, factual assertions do.
Write sentences that can stand alone. Include statistics with sources. Make claims that can be attributed.
Add citations and sources
Content that cites primary sources performs better in AI visibility. The original GEO research showed citation-heavy content saw up to 40% better visibility in AI responses.
Makes sense. AI systems are trying to provide accurate information. Well-sourced content is safer to cite.
Structure for extraction
AI parses content differently than humans. It's looking for direct answers.
Structure with:
- Clear question-and-answer formatting
- Concise summaries at section beginnings
- Factual statements that can be extracted verbatim
- Lists and structured data when appropriate
Build domain authority
AI systems don't cite random sources. They cite sources they "trust."
This is where backlinks from credible sources become critical. A link from Wikipedia, a citation in a respected publication, mentions on Reddit and Hacker News. These signals tell AI systems your content is worth citing.
Services like Revised exist for this. We find contextual backlink opportunities from sources AI systems already trust.
Go deep on topics
AI systems increasingly recognize topical expertise. A site that covers SEO deeply and consistently beats a generalist site with one SEO article.
Own your topics. Build content clusters that establish you as the authority in your space.
What hasn't changed
For all the AI disruption, some fundamentals stay constant.
Technical SEO still matters. If Google can't crawl your site, AI can't cite it. Core Web Vitals still impact rankings. Mobile performance still matters.
Content quality wins. AI just raised the bar on what "quality" means. But the core principle stays the same.
Backlinks still work. The mechanism is slightly different (authority for AI citation vs. PageRank for traditional ranking), but the outcome is the same.
User experience counts. Sites that satisfy users perform better. True in 2015, true in 2026.
A practical 2026 SEO framework
Here's what I'd do if I were starting fresh this year.
Audit your authority first
Where do you stand? Check your domain authority/rating in Moz or Ahrefs. Look at referring domains and their quality. See whether you're cited in AI responses for your key topics. Check how your content appears (or doesn't) in AI Overviews.
If signals are weak, prioritize authority-building before content expansion.
Double down on E-E-A-T
Add clear author bios with credentials. Include first-hand experience in content. Cite authoritative sources. Build external validation through earned media, backlinks, mentions.
Optimize for citation
Review your top content. Is it quotable? Does it include statistics and sources? Can key statements be extracted and attributed?
Rewrite for clarity and extractability where needed.
Build authority systematically
This isn't a one-time effort. You need ongoing content that earns natural links, active promotion, strategic backlink building from trusted sources, participation in industry conversations.
Track what matters
Add to your metrics: Are you appearing in AI Overviews? Are AI assistants citing your content? Is your referring domain count growing with quality sources?
Traditional rankings still matter. But citation visibility is the emerging metric to watch.
The bottom line
AI didn't kill SEO. It raised the stakes.
The sites winning in 2026 are the ones that would have won anyway. Credible, authoritative, genuinely valuable. AI just made it harder to fake.
If you've been doing SEO the right way, you're positioned well. Double down on what works.
If you've been cutting corners, the window is closing fast.
Ready to build the authority profile that gets you cited? See how Revised works.
More Articles You Might Like
Email Marketing for Small Business: A No-Fluff Starter Guide
I ignored email marketing for two years because it felt old-school. Then I watched a competitor's tiny newsletter outperform my entire social media strategy. Here's the starter guide I wish I had.
Online Marketing Strategies for Small Business (That Actually Work in 2026)
I burned $1,500 on Facebook ads and got three email signups. One was my mum. After six years of trying every channel as a bootstrapped founder, I can finally tell you which ones are worth your time.
Website Conversion Rate: How to Turn Traffic Into Customers
My first SaaS hit 5,000 monthly visitors and I celebrated. Then I checked the numbers. Eleven signups. Three trials. Zero paying customers. I had a traffic problem disguised as a success story.