Google Rolls Out March 2026 Core Update Targeting AI-Spam and E‑E‑A‑T

By Charles Ryder

If your rankings have felt like they’re shifting hour-to-hour since late February, you’re not imagining it. SEO tracking tools have been flashing volatility—Semrush Sensor climbing into the 9+ range, “winners and losers” changing daily, and entire SERPs rotating between homepages, product pages, and blog posts for the same query.

The community’s working name for this is the March 2026 Core Update. Whether Google officially labels it that way or not, the pattern looks familiar: a broad reweighting of quality signals, with a sharper focus on AI-spam and a renewed emphasis on E‑E‑A‑T (Experience, Expertise, Authoritativeness, Trust).

The twist is timing and visibility. Google has already said it’s shipping smaller, ongoing core changes without always announcing them. So this may not land as one clean, named event. It may show up as sustained turbulence that only becomes obvious once enough sites report impact.

The lead-up: from Discover to the main SERPs

February 2026 wasn’t quiet. On February 5, Google launched the first-ever Discover-specific core update. It rolled for more than three weeks and wrapped around February 27–28. Google’s stated goals were straightforward: elevate locally relevant stories, reduce clickbait and sensationalism, and reward deeper original reporting.

What matters now is what happened next. Volatility didn’t settle when Discover finished. It kept building, with spikes earlier in February and a noticeable heat-up as March began. That’s why many SEOs are treating early March as a separate broad adjustment—not a lingering Discover aftershock.

So far, Google’s public channels—Search Central, status dashboards, and the usual spokespeople—have been mostly quiet. That lines up with Google’s new posture: not every meaningful shift gets a headline.

What seems to be changing: AI content isn’t the target—AI spam is

Google has never said “AI-written content is banned.” The consistent message has been that automation is fine when it produces helpful content. What appears to be getting hit in March is the scaled, low-value end of the spectrum—the kind of output made to fill an index, not serve a reader.

Based on community analysis and the patterns showing up across sites, the likely targets include:

• Mass-generated pages that summarize other summaries (thin, recursive content)
• Auto-blogs and programmatic pages with no unique value
• Thin affiliate pages that add little beyond rewritten product specs
• Clickbait frameworks that win the click but disappoint on-page
• Low-trust niches leaning on volume over credibility (piracy/download patterns get mentioned often)

This doesn’t look like an “anti-AI” update. It looks like an anti-scaled-manipulation update—AI just happens to be the fastest way for spammers to scale.

Why E‑E‑A‑T is back at the center (especially the first “E”)

E‑E‑A‑T isn’t new, but the tone of March chatter is. The reported winners skew heavily toward sites showing first-hand experience and editorial accountability, not just decent writing.

What tends to perform better in this kind of reshuffle:

• Content written or reviewed by people with verifiable subject experience
• Pages with original insights, not just reassembled facts
• Strong topical coverage (clusters, not one-off keyword pages)
• Clear author pages, transparent policies, and consistent brand signals
• Helpful UX: fast, scannable, non-deceptive, and aligned with query intent

Local intent is part of this story, too. February’s Discover update already leaned into local relevance and depth. SMB-focused commentary is pointing businesses back to fundamentals like a complete Google Business Profile, strong on-site location signals, and mobile performance that doesn’t undermine trust.

The telltale sign: volatility that looks “broad,” not punitive

A common mistake during core update season is treating every drop like a penalty. Broad core updates are usually rebalancing events. Google re-evaluates what “best result” means for a class of queries, and some pages lose position because others now match the intent better.

That’s why we’re seeing odd rotations like:

• A homepage temporarily outranking the “right” article
• Product pages swapping positions with blog content
• Region-specific reshuffles (some Indian sites reporting early dips while other regions move later)

Those patterns typically point to system-level reassessment, not manual actions.

What I’m advising site owners to do (and not do) during the rollout

If this follows the familiar 2–3 week rhythm many broad cores do, the best move is simple: don’t panic-edit your site into a worse version of itself mid-stream.

Here’s the approach I’m using right now:

1) Confirm the problem is real (and where it’s happening)

• Compare GSC clicks/impressions by page type (informational vs. commercial)
• Segment by device (mobile vs. desktop mismatches are being reported)
• Figure out whether the loss is query-set specific or truly sitewide

2) Audit for “scaled sameness”

If you’ve published at high velocity, get blunt:

• Would I bookmark this page?
• Does it contain anything I couldn’t get from the top three results?
• Is it obvious this was written by someone who’s actually done the thing?

3) Upgrade experience signals

“Experience” doesn’t require oversharing. It does require evidence:

• First-hand photos, screenshots, test results, walkthroughs
• Real comparisons and tradeoffs (not generic pros/cons filler)
• Clear authorship and visible editorial standards

4) Tighten trust and usability

• Fix intrusive ads, misleading layouts, and thin above-the-fold sections
• Improve internal linking so topical depth is easy to see
• Keep Core Web Vitals and performance honest—especially on mobile

5) Don’t chase the SERP with gimmicks

Core updates are where shortcuts stop working. If your recovery plan is “publish 200 more AI posts,” you’re likely doubling down on the pattern Google is trying to demote.

What this means going forward: unannounced updates are the new normal

The bigger shift isn’t just that AI-spam is losing ground. It’s that Google is increasingly comfortable shipping impactful changes without giving the industry a neat banner to track. That pushes everyone toward a more disciplined approach:

• Build brands people search for by name
• Publish fewer pages, but make them meaningfully better
• Use AI as an assistant, not a factory—human editing and lived experience are the differentiators
• Watch volatility, but measure outcomes in weeks, not hours

Conclusion: Trust beats volume—especially now

The March 2026 volatility is another reminder that SEO keeps consolidating around one durable advantage: trust earned through real experience and useful, original work. If your strategy leans on scale, templates, or pages that exist mainly to “have a page for that keyword,” this is the moment to pivot.

If you want a practical way to keep an AI-assisted workflow aligned with E‑E‑A‑T—without drifting into the patterns Google is clearly suppressing—build around quality controls and real authorship. AIuthority helps you systemize human-first, trust-forward content instead of spam-at-scale.