Google’s AI Search Grows as AI Summaries Cut Outbound Clicks

The rapid financial collapse once predicted for Google Search has not appeared in the latest results. An Associated Press account of Alphabet’s second quarter found that revenue rose 24% year over year to $119.8 billion, with search-driven digital advertising helping propel the increase. That does not settle the long-term contest, but it shows that generative answers have not yet broken Google’s advertising engine.

What has changed is the shape of the threat. Google is no longer defending a conventional results page from a separate generation of chatbots; it has placed conversational answers, follow-up questions and agents inside Search itself. In its May 2026 Search update, Google said AI Mode had surpassed one billion monthly users, its query volume had more than doubled each quarter since launch, and total Search queries had reached a record high in the preceding quarter. These are Google’s own usage metrics, but they demonstrate that conversational retrieval can grow within its existing product rather than only at a rival’s expense.

Google absorbed the chatbot interface

The original competitive danger was easy to see. ChatGPT, Perplexity and similar services let people ask complete questions and receive synthesized responses, reducing the need to convert a problem into keywords, inspect several results and assemble an answer manually. If that behavior became the default elsewhere, Google could lose both the user’s attention and the commercial intent attached to the query.

Google’s response has been to make the distinction between search engine and chatbot less meaningful. AI Overviews place generated summaries above conventional results, while AI Mode supports longer prompts, follow-up exchanges and inputs such as text, images and files. The redesigned search box can expand for complex questions, and a user can continue from an AI Overview into an AI Mode conversation without leaving Search.

This approach protects Google’s position as the starting point. The company does not need users to prefer a page dominated by blue links if it can offer generated answers and web results in the same environment. Search also combines its web index with specialized information in areas such as shopping, maps, finance and sports, giving it assets that a general-purpose assistant must reproduce, license or access indirectly.

The billion-user figure does not represent a billion departures from conventional Google results. AI Mode is part of Google Search, so its adoption can reflect existing users changing interfaces rather than new users entering the service. Its significance is narrower but still substantial: chatbot-style retrieval is now a mass-market behavior inside the incumbent product.

The advertising crisis has been delayed, not disproved

Strong revenue weakens the simplest version of the existential-crisis argument—that AI answers would quickly drain search demand and advertising income. Advertisers still pay to reach people expressing commercial intent, and Google can introduce promotions into AI-assisted discovery just as it monetized earlier changes to search results.

Quarterly revenue nevertheless cannot reveal how profitable an AI response is relative to a conventional results page. The total can rise because of query growth, pricing, new advertising formats, stronger demand in valuable categories or performance elsewhere in the business. It is therefore evidence that the overall system remains commercially strong, not a controlled test proving that every AI interaction monetizes as efficiently as traditional search.

Generative search also creates costs and product tensions that ordinary revenue growth does not resolve. Producing an elaborate response requires model inference, while an agent that narrows a purchase or completes research may reduce the number of advertisements a user encounters. Google must preserve commercial opportunities without making its AI experience slower, less trustworthy or more cluttered than competing services.

The near-term problem is consequently one of execution rather than audience disappearance. Google must operate AI search economically, maintain answer quality and create advertising formats suited to conversational and agentic workflows. Rivals can still capture particular research, shopping or work sessions, but they are no longer competing against a company limited to the old search interface.

Websites face the more immediate loss

Google can retain a query and an advertiser even when the website supplying useful information receives no visit. That asymmetry appears in Pew Research Center’s browsing analysis of 68,879 Google searches associated with 900 U.S. adults in March 2025. Users clicked a traditional result after 8% of visits containing an AI summary, compared with 15% when no summary appeared; links cited within summaries received clicks after only 1% of visits. Sessions ended after 26% of pages with summaries, versus 16% of pages containing only traditional results.

The study has important boundaries. It examined a U.S. panel during one month, and researchers reconstructed the results pages in April because generated summaries can change. The findings cannot establish that every publisher, subject or market experiences the same effect, nor can observational browsing data prove that the summaries alone caused every difference in behavior.

Even with those limits, the comparison identifies a direct mechanism of disruption. A generated answer can satisfy the searcher before an external page opens. Citation may help an answer appear credible, but being listed as a supporting source is not equivalent to receiving the visit that funds advertising, subscriptions or affiliate revenue.

This separates Google’s position from that of publishers. Google can change the format of the results page while preserving the user relationship and selling access to intent. A website generally depends on an outbound click, a direct audience or another distribution channel; it cannot monetize attention that remains entirely inside the search interface.

The existential risk now extends beyond Google

Google still faces a genuine strategic threat. A competing assistant could become the preferred starting point for complex research, shopping or workplace tasks, while agents capable of completing transactions may bypass familiar search-ad formats. Poor answers, weak sourcing or intrusive monetization could also make users choose a service with a clearer experience.

Yet the available evidence does not show Google’s core business already collapsing. It shows an incumbent importing the challenger’s interface, building conversational use inside Search and continuing to generate advertising growth. The unresolved question is whether that adaptation transfers more of the economic damage to the websites beneath the answer.

If generated responses consistently satisfy queries without producing meaningful visits, publishers may have less revenue to support reporting, specialist analysis, reviews and reference material. That creates a dependency problem for every AI search provider: the systems need current, trustworthy material, but their interfaces can weaken the businesses that produce it. Google may survive the first stage of the chatbot challenge while intensifying a more fundamental crisis for the open web on which search depends.

Also read:

• Gemini Powers a New Era of Contextual Advertising in Google Search
• What is Search Intent And What is its Relationship with SEO?
• Google's AI Agents Integration in Search: 2026 Business Implications