For 25 years, Google has owned the search experience—10 blue links, lots of ads, and huge profits. Last quarter, the segment accounted for just over half of total revenue, and I estimate it contributed roughly 90% of profits.

But now, that profit center appears to be in the early stages of coming under pressure. AI chatbots are changing how users interact with search, offering single, direct answers instead of clickable links.

The clearest sign of this shift is the explosive growth in ChatGPT usage. In late 2023, OpenAI reported 100 million weekly users. Ten months later, that number doubled to 200 million, and just six months after that, it doubled again to 400 million. I estimate it will double once more to 800 million weekly users by the end of June—reaching that milestone in just four months. That translates to roughly 600 million daily active users, still well below Google’s estimated 2.1 to 2.5 billion daily users—but closing the gap fast.

The shift in consumer behavior is being driven by platforms like OpenAI, Perplexity, Grok, and Llama—building faster, smarter, AI-native search experiences. These products aren’t burdened by traditional ad models, allowing them to deliver what consumers increasingly want: a simple, direct answer.

If there’s one group outside of Google that understands the reality of traditional search, it’s Apple’s management team. We heard from one of them this week—Eddy Cue—during the Google antitrust trial, where he noted that Safari searches declined in April for the first time ever. That’s a strong signal that users are starting to search elsewhere. And it matters, because Google is the default search engine in Safari.

Reddit recently shared a similar observation: a drop in user growth they attributed to changes in Google Search traffic. Taken together, these signals suggest the shift is already underway.

A lesson I’ve learned over 30 years of researching and investing in tech is that it takes time for businesses built on habitual behavior to slow down. Yahoo, eBay, and Redbox are all examples that come to mind. Google is even more deeply woven into our daily lives—so much so that I’ve often referred to it as the fabric of the internet. That experience is one of the reasons why I believe investors should be paying close attention to the early signs of change.

These trends are starting to show up in the numbers. Google’s Search revenue was up 10% year-over-year in March, but signs point to a potential slowdown. If Safari volume declined in April—as Eddy Cue suggested—then Q2 results could come in weaker than expected. The Street is currently forecasting 9% growth in Search revenue for the June quarter, but that may be overly optimistic. If April was down sequentially from March, 5–7% growth is more realistic. In truth, any figure below 9% would likely be viewed as a material negative. And if the number comes in above expectations, investors may still use any resulting strength in GOOG shares as a selling opportunity.

The part I struggle with most is that I don’t know how Google gets out of this corner. If I’m right—that consumers want a simplified search experience—then by definition, the blue links go away. That would force Google to find new ways to monetize its 2 billion-plus daily users.

The company’s current approach, AI Overviews, effectively medicates the search results page with generative answers at the top. But the challenge is, that’s not what consumers want. They want a simple answer.

One possible path is for Google to shift from blue links to sponsored generative search. The problem with that approach is it risks eroding user trust—people just want the best answer, not the one someone paid for. That could lead to a decline in market share. Advertisers who currently appear organically as “the answer” are unlikely to pay for that placement, just as they don’t pay for organic links today.

Another option is that Google moves to a subscription model—say, charging $5 per month for search. In that scenario, if 100% of daily search users paid (which is highly unlikely), it would generate $140 billion in annual revenue. That compares to expectations of about $200 billion in Search revenue next year. In other words, this is a very tough challenge to solve.

It may sound like a joke, but I seriously recommend they ask Gemini for a path forward.

Disclaimer