Google’s May 2026 Core Update Reshapes AI Search Citations and GEO Strategies

I’ve been watching Google’s May 2026 Core Update closely, and this doesn’t look like a routine algorithm adjustment. Google framed it as a standard broad core update meant to surface “relevant, satisfying content,” but the timing points to something bigger. It arrived alongside Google I/O 2026, where Google made one thing clear: AI search is now central to Search, not a side feature. AI Mode has passed 1 billion monthly active users, AI Overviews have reached 2.5 billion, and Gemini 3.5 Flash now plays a core role in how answers are generated and cited.

That matters because this update is doing more than shifting rankings. It’s changing who gets cited, who shows up inside AI-generated answers, and who earns trust before a user ever clicks.

The real shift: rankings and citations are no longer the same game

For years, SEO strategy rested on a simple idea: rank higher, gain more visibility. That connection is weakening.

One of the clearest signals from the May 2026 update cycle is the growing split between traditional rankings and AI search citations. Industry research showed that top-10 Google rankings once predicted about 76% of AI Overview citations in mid-2025. By early 2026, that number had fallen to roughly 38%.

That is a major change.

In practice, a page can still rank well and lose visibility inside AI Overviews or AI Mode responses. Meanwhile, a brand with strong expertise signals, tightly structured passages, and consistent third-party validation can earn citations without holding the top organic spot.

That’s why GEO, or Generative Engine Optimization, has moved from an emerging idea to a necessary discipline. Traditional SEO still matters, but it now operates inside a broader visibility model.

Why Google’s guidance matters more than ever

In mid-May, Google published guidance on optimizing websites for generative AI features in Search. The company’s position was notable: there is no separate playbook for AI Mode or AI Overviews. According to Google, the same core signals that support strong rankings also support citations.

But the standard is clearly higher.

Google is rewarding original, first-hand, expert-led content more aggressively. That means identifiable authors, real credentials, transparent sourcing, technical cleanliness, and content that shows genuine experience instead of rephrasing what’s already out there.

This is where many brands will run into trouble. A lot of content programs were built for scale. AI search systems, by contrast, are looking for evidence: Who said this? Why should they be trusted? Where is the proof? Can the answer be extracted cleanly at the passage level?

If a page can’t answer those questions quickly, it becomes less useful to generative retrieval systems, even if it still performs reasonably well in classic search.

The timeline that made this update different

The May 2026 Core Update didn’t happen in isolation. It unfolded alongside a series of related product and ecosystem changes:

• Early in May, Google reportedly altered or removed some AI Overview click-tracking signals, deepening attribution blind spots.
• Around May 15, Google released its generative AI search optimization guidance.
• On May 20, Google I/O introduced major AI search expansion news, including user growth numbers and Gemini 3.5 Flash as the default model.
• On May 21, Google officially announced the May 2026 Core Update, with a rollout expected to take about two weeks.
• By May 29 and 30, industry observers reported strong ranking volatility, with some saying the update hit sites hard.
• By May 31, analysis across the search industry increasingly pointed to one conclusion: the update reinforced E-E-A-T, expert originality, and GEO-focused content strategy.

Taken together, this wasn’t just an algorithm refresh. It was a structural signal that Google’s AI-first search environment is now shaping how content is evaluated from the start.

What content is winning in AI search now

The pages most likely to earn citations in AI-generated answers tend to share a few traits.

1. They lead with the answer

AI retrieval systems favor extractable sections. Clear, concise answer blocks often perform better than pages that bury the point under long introductions. The strongest formats usually follow a BLUF approach: bottom line up front.

2. They are built for passage-level retrieval

Google’s generative systems can pull from specific sections rather than relying on the page as a whole. That makes structure critical. Tight 200–400 word sections, descriptive subheads, direct language, and focused topic clusters create more citable passages.

3. They show real expertise

Named authors, credentials, bios, photos, and direct experience are becoming essential trust signals. In sensitive sectors especially, anonymous or generic content is at a growing disadvantage.

4. They include verifiable proof

Facts, timestamps, source references, original data, expert quotes, and product-specific insights all help establish trust. AI systems appear to prefer content they can ground confidently.

5. They align across the web

Google’s broader research and third-party analysis point to the value of multi-source consistency. If your claims are reinforced across your website, industry publications, comparison pages, community discussions, and other credible sources, your chances of citation improve.

That last point matters a lot. AI visibility is not just about what your site says about itself. It’s about whether the rest of the web supports the same picture.

E-E-A-T is no longer a vague concept

We’ve talked about E-E-A-T for years, but this update makes it much more operational.

Experience, Expertise, Authoritativeness, and Trustworthiness are no longer just quality-rater concepts in the background. In the generative search era, they increasingly determine whether your content can be used as a source.

That means brands need to move beyond generic educational content and invest in material that reflects real subject-matter authority. Thoughtful commentary, first-hand use cases, product-specific insights, original research, implementation lessons, and expert-authored analysis all carry more weight now.

I’d go a step further: first-hand experience is becoming a competitive moat. It’s harder to imitate, easier to trust, and more valuable to both users and AI systems.

Why traffic data alone is becoming misleading

One of the more frustrating side effects of the May 2026 update is measurement.

As AI answers become more prominent, many brands are seeing stable rankings but weaker click-through rates. Top-ranking pages can lose significant CTR when AI Overviews satisfy part of the query upfront. At the same time, AI visibility may still shape conversion paths even when the click doesn’t happen immediately.

That creates a serious attribution problem.

If a buyer discovers a brand through an AI Overview, returns later through direct traffic, and converts after a branded search, the original influence can disappear from standard reporting. With Search Console and analytics still offering limited clarity into AI Overview interactions, marketers need a new measurement model.

I think the smartest teams will start tracking:

• citation rate across priority queries
• share of voice in AI-generated answers
• AI-referred sessions where visible
• self-reported attribution from leads
• CRM segmentation tied to branded lift and assisted conversions

The old formula of rankings plus organic traffic equaling SEO success no longer tells the full story.

GEO strategy is now a three-part discipline

The most useful framework emerging right now is the idea that brands need to win three games at once:

Ranking

Traditional SEO still matters for discoverability, crawlability, and sustained organic traffic.

Citation

Your content must be structured and trustworthy enough to be selected inside AI answers.

Agent-source trust

As search becomes more agentic, systems will increasingly evaluate whether your brand is consistently validated across sources, formats, and contexts.

That third layer is where many strategies are still thin. Publishing a strong page is no longer enough. Your expertise has to be corroborated across the wider web.

What marketers should do next

If I were advising teams right now, I’d focus on five immediate actions.

Audit key pages for citation readiness

Review your highest-value commercial and informational pages. Are they easy for an AI system to quote? Do they offer direct answers, clear subheads, strong evidence, and distinct expertise?

Add visible expert signals

Every important page should show who created it, why they’re qualified, and when it was updated. If expertise exists behind the scenes but not on the page, it’s being wasted.

Strengthen original content

Replace commodity summaries with first-hand insights, examples, research, opinions, and operational knowledge. If your page sounds like everything else in the category, it’s easy to overlook.

Improve cross-source consistency

Review how your brand, product claims, and category positioning appear across third-party sites, review platforms, media mentions, partner content, and community discussions.

Update measurement frameworks

Start treating AI citation visibility as a business metric, not just an SEO curiosity. Build attribution systems that capture influence before the click.

The future belongs to citable brands

The biggest lesson from Google’s May 2026 Core Update is simple: search visibility is no longer just about where you rank. It’s about whether your content is trustworthy enough, clear enough, and useful enough to be synthesized into answers.

That is a higher standard, but it’s also a better one.

Brands that invest in real expertise, strong structure, and cross-web trust will benefit. Brands that keep publishing interchangeable content at scale will become increasingly invisible, even if their rankings still look decent on the surface.

I see this update as a dividing line. On one side are teams still optimizing for the old search landscape. On the other are teams adapting to a world where SEO, GEO, and AI citation strategy are inseparable. If you want help building content that earns visibility in that environment, I recommend exploring AIuthority to make the shift faster and more strategically.

FAQ

What changed in Google’s May 2026 Core Update?

The update appears to reinforce a broader shift toward AI-first search. It affected traditional rankings, but more importantly, it changed how content is selected for AI citations in AI Overviews and AI Mode.

What is GEO?

GEO stands for Generative Engine Optimization. It focuses on improving how content is cited, summarized, and surfaced in AI-generated search results, not just how it ranks in classic organic listings.

Can a page rank well and still lose AI visibility?

Yes. That is one of the clearest patterns emerging from this update. A page can hold strong organic rankings and still fail to earn citations if it lacks structure, proof, expertise signals, or passage-level clarity.

What types of content are more likely to be cited?

Content that leads with direct answers, uses clear structure, demonstrates real expertise, includes verifiable proof, and aligns with credible sources across the web is more likely to be cited.

How should marketers measure success now?

Marketers should look beyond rankings and traffic alone. Citation rate, AI share of voice, assisted conversions, branded lift, and self-reported attribution all matter more in an AI-driven search environment.

Conclusion

Google’s May 2026 Core Update marks a clear turning point. Search is no longer just a ranking contest; it’s a trust and citation contest as well. The brands that win next will be the ones that publish original, expert-led content, structure it for retrieval, and reinforce their authority across the broader web.