In October 2025, Sam Altman quietly let the cat out of the bag on stage at OpenAI DevDay: ChatGPT had hit 800 million weekly active users, doubling from 400 million in February of the same year (TechCrunch, Oct 6, 2025). By February 2026, that number had climbed past 900 million (Backlinko ChatGPT Stats).

For context: it took Google ten years to reach a billion users. ChatGPT got there in roughly three.

This is not a “rising tide” story. It is a tectonic shift in how human beings find information - and most brands are still optimizing for a search behavior that is collapsing under their feet.

The 90% line just broke

Google’s global search market share dropped below 90% in late 2024 for the first time since 2015, according to Statcounter data (Search Engine Land, 2024). It is the first sustained crack in a fifteen-year monopoly, and the cause is not Bing or DuckDuckGo. It is a category that didn’t exist in 2022.

Gartner’s now-famous February 2024 forecast put hard numbers on the trajectory:

“Traditional search engine volume will drop 25% by 2026, with search marketing losing market share to AI chatbots and other virtual agents.” - Gartner, February 19, 2024

A 25% drop in traditional search volume is not “a slowdown.” It is the largest reversal in consumer search behavior since the launch of mobile.

Where the queries are going

ChatGPT is the headline number, but the broader market tells the same story.

Perplexity - the citations-first AI search engine - processed roughly 780 million queries in May 2025, growing 20% month-over-month, with the platform now handling around 30 million queries per day per CEO Aravind Srinivas (Just Think AI, 2025).

Google’s own AI Overviews now appear on roughly 13.14% of all queries, up from 6.49% in January 2025 - and McKinsey projects that share will exceed 75% of all Google searches by 2028 (McKinsey, “New Front Door to the Internet,” 2025). Even when users stay on Google, the experience is being rebuilt as a generative answer engine.

Consumers have already shifted

The behavior data is even more decisive than the platform numbers. McKinsey’s 2025 report on AI search found:

  • ~50% of consumers now intentionally use AI-powered search.
  • 44% say AI is their primary insight source (vs. 31% for traditional search).
  • McKinsey projects $750B in US revenue will route through AI-powered search by 2028.
  • Unprepared brands face a 20–50% decline in traditional search traffic over the same window.

Bain’s Consumer Lab GenAI Survey (December 2024) adds the shopping dimension:

  • 30–45% of US consumers use generative AI for product research and comparison.
  • 51% trust GenAI for shopping recommendations.
  • 52% of Millennials and 25% of Gen Z plan to start holiday shopping with an AI assistant.

Pew Research confirms it generationally: 34% of all US adults have used ChatGPT as of early 2025 - double the 2023 figure - and 58% of adults under 30 have used it (Pew Research, June 2025). 31% of Americans now interact with AI several times daily, up from 22% in February 2024.

The buyer your brand needs to reach in 2026 isn’t typing keywords into a 10-blue-links page. They’re asking ChatGPT a conversational question and trusting the answer.

Why “AI search” is structurally different

Three things break in the move from traditional SERPs to AI agents:

1. The result is a single answer, not a list

Google gave you ten links. ChatGPT gives you one paragraph. There is no “page two.” If your brand isn’t named in the answer, you don’t exist in that buying decision.

2. The user doesn’t see your site

In traditional SEO, a high rank produces a click. In AI search, a citation produces a brand mention - often without any click at all. The metric that mattered (CTR) is being replaced by one that didn’t exist (cite-through).

3. The retrieval surface is different

Google indexed your pages by link graph and keyword relevance. AI agents retrieve via embeddings, schema, entity authority, and structured citation patterns. The optimization surface has changed entirely.

This is the gap that Generative Engine Optimization (GEO) exists to close.

The publisher canaries are already dead

If you want a leading indicator, look at publishers - the industry most exposed to the search-to-AI shift.

  • Press Gazette: Global publisher Google traffic dropped ~33% from November 2024 to November 2025; US publishers down 38% (Press Gazette, 2026).
  • Chartbeat: Small publishers lost 60% of search referral traffic over two years; medium publishers lost 47% (Axios, March 2026).
  • News publishers themselves expect search referrals to drop 43% by 2029 (Search Engine Land).

These aren’t speculative declines. They are observed, measured drops in traffic from the same audience now getting their answers from AI summaries.

What this means for brands

If your category is being searched, your category is being asked. The question is who gets named in the answer.

Three things become non-negotiable:

  1. You must be retrievable. That means clean technical infrastructure, schema markup, fast crawl surface, and structured content that LLMs can parse and quote.
  2. You must be cite-worthy. That means authoritative voice, statistics, original data, and the kind of definitive language LLMs prefer to extract.
  3. You must be measured. Not in Google rank - in cite rate across every major agent: ChatGPT, Claude, Perplexity, Gemini, Grok, Copilot.

The brands winning AI visibility today are not the brands with the biggest SEO budgets. They are the brands engineered, from the schema up, to be the answer.

Sources cited in this article: