If Google Search is turning into Gemini, what’s the point of Gemini?

For the past couple of years, Google I/O has been the stage exclusively for AI announcements, and this year wasn’t any different. This time, however, the focus was back on Google Search, but perhaps not in the way we expected. Instead of addressing the long-standing, fundamental problems with Search, Google introduced a lot of AI integrations into its marquee search engine — so much so that the line between Google Search and Gemini is increasingly becoming difficult to make out.

While AI Overviews and AI Mode have been meaningful additions to Search despite their rough start, it now seems that Google is merging two products with fundamentally different purposes. That makes one wonder: if Search can now behave like Gemini, what exactly is Gemini supposed to be?

Artificial intelligence has fundamentally changed how we interact with search engines. A lot of us have already started being more conversational in our queries, typing longer questions, explaining our perspective, using natural language, and asking follow-up questions. The new unified search box — that brings together the simplicity of conventional search and the reasoning of AI — further reduces friction because users no longer have to think about using the right keywords in searches.

Search’s primary input method has been text, but we have long been able to reverse-search images quite successfully. The next logical step is giving it multimodal capability to help it understand and search based on different media types like video and audio, and combine them with your text input. And it’s finally getting smarter at keeping track of price drops and new product launches in the background and alerting you as soon as it sees something move. 

These kinds of information agents in Search are indeed useful, as long as you are fine with braving another notification avalanche from your search engine. This feature, specifically, sits right at the boundary that differentiates Search from Gemini. 

Search is supposed to have a very specific function: looking up the web and bringing up direct links that match the query. Gemini, on the other hand, also has access to the entire World Wide Web and all the information that resides within it, but its job goes a couple of steps further — understanding that information and explaining it to me in a way I understand. AI Overviews and AI Mode partially took over that job within Search, but Gemini’s separate identity remained unaffected through its multimodal capabilities, embedded generative tools, and Google Workspace-wide integration. 

Only now, Google Search is inheriting quite a bit of those generative skills. Besides letting you ask follow-up questions, Search has now borrowed agentic coding abilities from Gemini, allowing it to create interactive elements from scratch to help you understand a topic better. Google didn’t stop there. It went a step ahead and introduced the ability to create stateful mini apps inside Search, letting you create dynamic layouts, dashboards, and interactive widgets for long-running projects like home makeovers or wedding planning. 

Helpful, sure, but Google Search is no longer fetching information for you. It is instead creating things from scratch, setting up workflows, and managing persistent projects. When I think of these capabilities, my mind gravitates towards using Gemini and not Search. If I want a complex topic explained through interactive tools and elements, Search would honestly be the last place I’d expect to look for something like this. 

This disorderly implementation creates an overlap between two of Google’s most-used products. It’s almost as if Google is inadvertently making Search its real AI assistant, despite Gemini existing as its own app and brand. 

We understand their respective purposes: Search has forever been the tool for finding new stuff and getting quick answers, whereas Gemini is more involved, with things like planning, reasoning, generating, and executing. I frequently use Gemini to replace some mundane tasks in my workflow. While I understand that AI would naturally blur some boundaries, that shouldn’t change the fundamental role of each product, and currently, it appears that Google is pushing both Search and Gemini in the same direction, with the same end goal. 

I came out confused from the Google I/O presentation, and I am sure that a lot of users will be too. Right now, if my mind buzzes with a question, I mostly know whether it is meant for Search or Gemini, depending on the kind of output I want. If I’m looking to buy a pair of sneakers that I spotted somewhere, I’d go to Search to find the brand’s website. If I wanted to understand what makes sneakers so comfortable, then I’d direct my query to Gemini. With the lines blurring, I’d spend more time wondering which tool would be better for the task at hand than actually interacting with it and getting my answers before moving on. 

From my vantage point, Google is grossly ignoring a branding advantage it has. Instead of letting Search do all the generative work, why not hand those tasks to Gemini? This way, Search does what it was made for — discovery — while Gemini manages the planning and execution part. This middle ground doesn’t dilute either product’s individual identity and also saves us from remembering tens of tool names within Search — AI Mode, AI Overviews, et al. 

Having covered Google for so long, I already know that it’s not going to make things as straightforward, at least not without first hitting a roadblock itself. But I’ll still ask away. 

When Google has been pushing the Gemini branding for the past couple of years at literally every single place, while also essentially turning its most popular product — Search — into Gemini, why not just fully commit to that transition and call it Gemini Search? And if Search is able to handle everything that Gemini can — which it evidently now does — then what was the point of a separate Gemini branding in the first place? 

Google Search has transformed in the last few years almost as fast as AI did — and that speed has not only magnified its existing issues but also introduced new ones. Google Search these days is riddled with SEO-optimized spam and AI-generated content that has reduced trust in result quality so much that we often resort to appending our queries with “Reddit” just to get real answers from actual humans. 

It wasn’t too long ago that AI outputs were severely criticized for creating an utter mess in their summaries. Some of them even potentially endangered human lives by equating parachutes with simple backpacks. While those problems have subsided considerably, Google is now turning Search into a full-fledged AI platform at full speed. Instead of fixing the experience of what is quite literally the storefront of the internet, Google is simply layering it with AI, hoping you don’t notice the crumbling foundation underneath.

It’s quite evident that Google wants AI to be everywhere, but the real problem is that Gemini and Search now look like two interpretations of the same core idea. And with that overlap only increasing from here, it’s becoming harder to understand why Google wants both of them to exist separately at all. Google should either kill one of these services or let them live up to their own individual potential — I don’t see any other way out.