Alphabet Inc. (NASDAQ: GOOGL, GOOG) has a problem most companies would love to have. Its search engine is so profitable that it can pay for one of the largest AI build-outs in history. The catch is that the same AI is changing how people search, and how much Google earns each time they do.
In October 2026, Google said searches were growing at their fastest pace in five years [1]. Yet search ad growth slowed, and the company linked part of the slowdown to its chat-style AI Mode, which it said earns less from ads for now [2]. The shares fell sharply after the report.
When investors ask whether AI will replace Google search, three questions tend to follow. Can AI search earn as much as classic search? Will the huge AI spending earn a fair return? And can cash flow recover while that spending continues?
This guide explains how Google makes money, where AI helps and where it hurts, and the signals worth tracking every quarter.
#How Google Makes Money, and Why Search Still Pays the Bills
Most of Alphabet's money still comes from advertising. In the quarter to September 2026, ads brought in about two-thirds of total revenue, and search ads alone were close to half.
The model is simple. Businesses bid to appear when someone types a question with buying intent, such as "best travel insurance." Google is paid mainly when people click. Searches linked to shopping, travel, insurance and finance are the most valuable, because advertisers pay more to reach someone ready to spend.
Google also pays partners to put its search engine in front of users, for example as the default on phones and browsers. These payments are called traffic acquisition costs, or TAC. They are a large expense, and they sit at the heart of the antitrust cases covered below.
Google Cloud is now the second big engine. It sells computing power, AI tools and Google-designed AI chips to businesses. Cloud made up roughly a quarter of revenue in the same quarter, after more than doubling from a year earlier. YouTube, subscriptions and devices make up most of the rest.
#The Bigger Risk Is Lower Revenue per Search
So far, AI has not emptied Google's search box. Google said query growth in 2026 was the fastest in five years, helped by its AI features. It still handled about 86% of US search engine traffic in mid-2026, though that measure does not count chatbots such as ChatGPT [3].
The risk sits in how AI search is used. Google now runs three different AI experiences. AI Overviews are short AI summaries shown above normal results. AI Mode is a chat-style search tab for longer questions. Gemini is a separate AI assistant app.
Each one changes the money math. A classic search shows a page of links and ads. An AI conversation can replace several searches, and each of those searches was a chance to show an ad. If one long chat earns less than the five searches it replaced, Google can serve more people while earning less from each question.
Google's executives pointed to this effect in October 2026. They said AI Mode was growing fast but earning less from ads than traditional search, and that ads in AI Mode were being rolled out to more countries. One portfolio manager said investors needed to see search growth speed up again before treating AI as a clear plus.
AI answers also change where users go next. In one study of US browsing, people clicked a traditional search link in 8% of visits when an AI summary appeared, compared with 15% when none appeared [4]. Fewer clicks hurt publishers, review sites and other businesses that rely on Google traffic. That matters to Alphabet too. Its AI answers depend on a healthy web to summarize, and many advertisers judge Google by the visits it sends them.
Ads inside AI answers also raise a trust question. Google says ads in AI Mode carry a “Sponsored” label and do not change the core answer [5]. Its own pitch to advertisers describes newer formats built directly into AI responses that explain why a product is the best answer to a question [6]. Critics argue that ads placed mid-conversation can shape how people read an answer, even when the answer itself is unchanged. For investors, this cuts both ways. Ads that blend in may earn more, but if users or regulators start to doubt the answers, usage and rules could both shift.
A sharper way to see the markets in just 5 minutes.
Same news, different lens. We cut through the noise and hand you the overlooked ideas and the deeper read the crowd misses. Join 38,000+ investors seeing the markets differently.
#Alphabet's AI Spending Puts Free Cash Flow to the Test
Building AI takes data centers, chips, networking gear and huge amounts of electricity. This spending is called capital expenditure, or capex. The cash leaves today, while any profit it produces arrives later.
Free cash flow is the cash a business has left after paying for capex. For years, Alphabet produced a lot of it. That changed in 2026. Free cash flow turned negative in the second and third quarters, because capex grew faster than the cash the business brought in. Management has signaled that spending will keep rising.
The way Alphabet funds this matters too. During 2026 it raised money by selling bonds and issuing new shares, and it made no share buybacks in the first nine months of the year, after spending tens of billions on them a year earlier. New shares can dilute existing holders, meaning each share owns a slightly smaller slice of the company.
Three slower-burning risks sit behind the headline spending. Servers and chips are written off over several years, so today's capex becomes a cost against profit later, a charge called depreciation. AI hardware improves fast, so some equipment could lose value sooner than planned. Power shortages and construction delays could also raise costs.
The counterpoint is scale. Alphabet designs much of its own AI hardware and runs some of the world's largest ad and cloud businesses. If demand for AI keeps growing, that position could help it earn a solid return. The open question is how quickly.
#Google Cloud Gives Alphabet a Second Growth Engine
If search is the cash machine, Cloud is the part of Alphabet that sells AI to other businesses. It rents out computing power, offers AI models and tools, and now sells its in-house AI chips, called TPUs, directly to customers.
Cloud has grown far faster than the rest of the company and has become much more profitable. In the third quarter of 2026, its operating profit roughly tripled from a year earlier. Google also reported a backlog, meaning business already signed but not yet delivered, of about $561 billion.
Cloud carries its own risks. Much of the AI demand comes from a small group of very large customers, so losing one could hurt. Google has also rented extra computing capacity from other providers to keep up, which management said could squeeze margins in the near term. And Cloud competes head-on with Amazon Web Services and Microsoft Azure.
Other businesses could help spread the risk over time. These include YouTube, paid subscriptions such as YouTube Premium and Google One, and the Gemini app, which Google said had reached one billion users. Waymo, the self-driving car unit, sits in a group of early-stage projects that together still lose money.
#Competition and Antitrust Rulings Could Weaken Google's Defaults
Google faces rivals on two fronts. AI assistants from OpenAI, Microsoft, Anthropic, Perplexity and others compete for people's questions. Social media and shopping platforms compete for product discovery, especially among younger users.
Regulators are the second pressure. US courts have found that Google illegally protected monopolies in search and in parts of its advertising technology [7]. So far, judges have chosen new rules over breakups. In the search case, Google was not forced to sell its Chrome browser or Android, but was ordered to share some search data with rivals and to end some exclusive default deals [8]. In the ad tech case, a judge declined to break up the business in September 2026 but ordered changes to how it operates, overseen by an internal antitrust compliance monitor [9]. Both Google and the government have appealed parts of the search ruling.
Why does this matter to investors? Being the default search engine is one of Google's strongest advantages. Rules that weaken those deals, or hand useful data to competitors, would more likely wear down its position over years than months. That slow pace is also why the outcome is hard to price.
#Weighing Google Stock Risks Against the Bull Case
Investors read the same facts in very different ways. The table sums up the main arguments on each side. Neither column is a prediction.
Theme | What supporters argue | What skeptics argue |
|---|---|---|
AI search | AI brings more searches and new ad formats | AI answers earn less per question and send fewer clicks |
AI spending | Own chips and huge scale could make AI cheaper to run | Spending could outrun returns and keep cash flow weak |
Cloud | Fast growth and a large backlog spread the risk beyond ads | Growth leans on a few big AI customers and costly capacity |
New products | Gemini, subscriptions and YouTube add recurring income | Competition could limit what Google can charge |
Competitive moat | Scale, data and distribution are hard to copy | Courts and rivals could weaken default deals and data advantages |
#Who Else Is Exposed to Google's AI Bet
Alphabet's choices ripple well beyond its own shareholders, across companies of every size.
Group | Why Google's direction matters to them |
|---|---|
Advertisers and agencies | AI search changes ad formats, costs and how results are measured |
Publishers, review and comparison sites | Fewer clicks can mean less traffic and less ad income |
Chip, memory and networking suppliers | Alphabet's capex is a large source of their sales |
Data center builders and utilities | AI sites need land, power and water |
Rival cloud and AI firms | Google's pricing and chip sales shape competition |
Index fund holders | Alphabet is one of the largest companies in major US indexes |
#What to Watch in Every Alphabet Earnings Report
These signals tend to say more about the long-term story than any single quarter's profit figure. Most appear in Alphabet's quarterly earnings release, its 10-Q filing or on the earnings call.
Signal | Why it matters | Where to look |
|---|---|---|
Search & other revenue growth | Shows whether AI is adding to or eating into the core business | Revenue table in the earnings release |
Comments on ads in AI Mode and AI Overviews | Shows whether AI answers can earn as much as classic search | Earnings call |
Paid clicks and cost per click | Separates more ads from pricier ads | Quarterly 10-Q filing |
Capex and spending plans | Shows the size and length of the AI build-out | Cash flow statement and earnings call |
Free cash flow | Shows whether the business is funding itself again | Earnings release |
Depreciation | Shows today's spending turning into tomorrow's costs | Cash flow statement |
Cloud revenue and operating income | Tests whether fast growth is turning into profit | Segment results in the earnings release |
Cloud backlog | Shows business already signed for future years | Earnings call and 10-Q |
Traffic acquisition costs (TAC) | Shows the cost of staying the default search engine | Supplemental table in the earnings release and Cost of Revenues table in the 10-Q |
Share count, buybacks and new share sales | Shows dilution and how much cash goes back to holders | Financial statements |
Legal and regulatory rulings | Could change default deals, data sharing or ad tech | Court rulings and company filings |
#Frequently Asked Questions About Google and AI
Is AI killing Google Search? Not on current evidence. Google reported its fastest search query growth in five years in 2026. The worry centers on earnings, because AI answers may earn less from ads than classic searches.
What is Google AI Mode? It is a chat-style tab inside Google Search that answers longer questions as a conversation. Google is still testing ads inside it and has been expanding them to more countries.
What is the difference between AI Overviews, AI Mode and Gemini? AI Overviews are short summaries shown above normal search results. AI Mode is a conversational search tab. Gemini is a separate AI assistant app. Each may make money in a different way.
Do ads in Google AI Mode change the answers? Google says no. Ads carry a “Sponsored” label, and the company says they do not influence the core answer. Critics say ads placed inside a response can still shape how people read it.
Why does Alphabet's capex matter to shareholders? Capex is cash spent today on data centers and chips. When it grows faster than the cash the business brings in, free cash flow falls, as it did in mid-2026. Investors then watch whether that spending earns a fair return.
What is the difference between GOOGL and GOOG? Both are Alphabet shares. GOOGL is Class A stock and GOOG is Class C stock. Class A shares carry voting rights, while Class C shares do not. They usually trade at similar prices.
Does Alphabet pay a dividend? Yes. In October 2026 it declared a quarterly cash dividend on its Class A, B and C shares. Dividends can change and are never guaranteed.
How could Google's antitrust cases affect the business? Courts have so far ordered changes in how Google operates rather than breakups. Limits on default deals and required data sharing could help rivals over time.