Google is integrating more AI into search, including a dynamic search box and agent features

Google Search is becoming even more of a centerpiece of its AI ambitions. Today at Google I/O 2026, the company announced that Search has been upgraded to the Gemini 3.5 Flash model, which it says offers faster inference, smarter results, and the ability to process different media types. As a result, Google is also launching a new smart search box that can dynamically expand to accommodate complex queries, as well as use videos, images, files, and even Chrome tabs as inputs. However, none of this comes as a surprise.

Over the past year, we’ve seen Google gradually upgrade AI mode in search. Today, the company is simply making AI a more essential part of web search. That’s good news for users who value the more conversational capabilities of Google’s AI, but it’s even worse news for those trying to maintain the simple purity of Google’s original search engine. It’s unclear if you’ll be able to disable any of the new AI features – it was already difficult to get rid of the old AI search features.

The new Google search features, which the company says are the most significant changes to the search field in 25 years, are available today. “It’s more intuitive than ever, dynamically expanding to give you the space you need to describe exactly what you need,” Liz Reid, Google’s head of search, said in a blog post. “Designed to anticipate your intent, it also helps you frame your question with AI-powered suggestions that go beyond autocomplete.”

As you might expect, your AI queries won’t be as fleeting as old-fashioned search prompts. The company says you’ll be able to continue asking questions after your initial prompts, and that should lead to more accurate results. We’ll have to take a closer look at Google’s revamped search to see if the company’s claims are true, but at first glance it seems like it could be a clever way to get users to spend more time with its AI tools. This is exactly what Google wants, of course.

For its paid Gemini Pro and Ultra users, the company also plans to launch a series of new agent features this summer. Like other agentic AI features, these are meant to run on their own after an initial prompt. Future information agents, for example, should be able to scour the web for specific apartments after you tell Search what you’re looking for. Agents will be able to see “everything” on the web, according to Reid, including “blogs, news sites and social posts, as well as our most recent data, such as real-time financial, shopping and sports information.”

In some ways, the new agentic search is like an improved version of Google Alerts, which is one of the easiest ways to track keywords appearing on the web over the past two decades. The big difference now is that you’re looking for more complex information than a simple string of text, and Google’s AI should be able to format its results in a more useful way.

Free users will also benefit from a handful of agent updates over the summer, including the ability to book “local experiences and services.” The company says you’ll be able to search for something like a karaoke room, find accurate rates, and finalize your reservation. You can also ask Google to call certain businesses for appointments, similar to the Duplex AI feature launched with Google Assistant in 2018. Let’s just hope it’s less jarring – Duplex AI seemed way too robotic and it could be confusing for local business owners who have no idea why a Google bot is calling them.

The company also plans to introduce its agent coding application, Antigravity, into research this summer. According to Google, Antigravity will create generative UI elements to answer questions, like understanding how your watch works. Even more intriguing, the company says you can use Antigravity to create “mini apps” within Search, which could, for example, create a personalized fitness tracker just for you. This feature will initially come to Google AI Pro and Ultra users in the United States “in the coming months.

Clearly, Google’s goal is to make search a centerpiece of your life, instead of just a tool you use briefly to find other websites. We’ve already seen this in action with existing AI summaries, which summarize information without even needing to visit another site. Again, I haven’t tested any of the new search features, so I can’t really judge whether they meet Google’s expectations. But introducing Antigravity to research could be an easy way to get users used to the idea of ​​generative coding. It won’t be for everyone, but I could see some power users looking into it.

The company also claims to extend its AI-based personal intelligence to nearly 200 countries and 98 languages, without the need for a subscription. The feature allows you to connect Gemini to your Gmail, Photos and Calendar information, with the aim of providing more accurate results. The company emphasizes that “you are always in control” of these features, since you can specifically choose to connect your accounts. But given how much Google wants to take the lead in consumer AI, I wouldn’t be surprised if the company started enabling all your accounts by default.