Google has announced an AI-driven overhaul of Search centred on a redesigned “intelligent search box” — which the company claims is the biggest change to this web entry point since the search box first launched over 25 years ago.
Google Search occasionally launches users into AI-powered interactive experiences rather than just a list of links. Along with tools that enable users to create customised small apps based on their needs, Google is also launching tools that may send out “information agents” to collect data on a user’s behalf.
Google Search, which has long been characterised by ranking links to websites that contain the information you need, will no longer resemble the final experience.
Instead of requiring you to select a search experience or mode at the beginning of your inquiry, the new search field just extends to allow longer, more conversational searches with the redesigned search experience.
Per Google, it will also have a new AI-powered query recommendation engine that goes beyond autocomplete to assist users in crafting more intricate and subtle questions.
According to the firm, starting this week, users will be able to ask follow-up questions in AI Mode with Google’s AI Overviews.
Additionally, Google is enhancing the search experience with AI-powered interactive elements and agentic capabilities. As a result, users will click on the conventional blue links that Google Search used to return even less frequently.
People will be able to develop, modify and oversee several new “information agents” within Google Search starting this summer. These agents can monitor changes on the web and notify you of new information while operating in the background around-the-clock. For example, Google recommends that you have an agent monitor changes in the market depending on consumer characteristics.
Although AI powers the underlying technology, making it more capable, the concept itself is not new.
In 2023, Google introduced Google Alerts, a change-detection tool that sent customers an email whenever their search keywords appeared in new online results.
Naturally, the web was more controllable and smaller back then, thus many information professionals included this in their toolkits. (That service is still available in some capacity, although most web users no longer utilise it to find new material.)
Google Alerts have evolved into information-gathering agents. They are able to not only identify changes but also interpret them.
In a presser, Liz Reid, Google’s head of search, said, “You could send an alert to track market movements in a particular sector with very specific parameters, and the agent will map out a monitoring plan for you, including the tools and the data it needs to access – like our real-time finance data.” “And it will then monitor those modifications, notify you when the requirements are fulfilled, and offer a condensed update with links and details you can explore further,” she continued.
This change implies that AI agents will increasingly conduct “searching the web” instead of people. Instead of clicking links by hand, consumers will concentrate more on taking action based on the information those agents supply.
The upcoming modifications to the search results experience, which expand on Google’s previous introductions of AI search capabilities like its conversational search, AI Mode and its brief summaries, AI Overviews, will make links an afterthought.
More than 2.5 billion people use AI Overviews each month, while more than 1 billion people utilise its conversational search option, which was introduced last year.
In contrast, as of earlier this year, ChatGPT had 900 million weekly active users. This implies that while Google has more unique users interacting with its AI capabilities over the course of a month, ChatGPT is currently experiencing more regular interaction, with users returning frequently throughout the week.
Gemini and Google Antigravity, the company’s agentic development platform, are now working together to make search results more engaging.
According to Reid, “Search can create unique experiences just for your specific questions, from dynamic layouts, interactive visuals, to persistent and stateful project spaces that you can return to repeatedly.” Google’s “generative UI” (user interface), which creates unique widgets and visualisations in response to users’ search queries, is one method of incorporating these new features.
For instance, an inquiry about black holes in space can result in an interactive graphic that makes the idea come to life, according to Reid. Users can then ask follow-up enquiries, and Google would instantly answer with fresh visuals.
According to Google, Gemini Flash 3.5 is used in the new system, which was developed in collaboration with the Google DeepMind team. This summer, it will be made available to all Google users at no cost.
Additionally, Google will let users utilise Antigravity to create their own stateful, customisable experiences—think of them as “mini apps”—directly in Search by utilising natural language commands.
Moreover, action is more important here than information retrieval. For example, you might make a workout app tailored to your particular objectives or a meal-planning app that uses data from your personal calendar to assist you choose what to prepare and when to eat.
When taken as a whole, these modifications would probably significantly reduce Google’s referrals to publishers, who have already been experiencing a decline in referrals as a result of AI Overviews. Some ad-dependent media businesses have been forced out of business as a result, and things will probably grow worse from here on out.
There is not much time left for publishers to change. This week will see the arrival of the new search box, and this summer will see the arrival of generative user interface. They are both free.
This summer, Google AI Pro and Ultra subscribers will have the first access to the information agents and mini-app-building functionality.
However, Google’s long-term goal is to make its AI technologies more widely available, including Spark, a personal AI agent that will someday be free, along with many other AI capabilities.
”I believe that’s an area where we will shine,” Google CEO Sundar Pichai stated in a press conference prior to I/O. “Part of the reason we focus on delivering frontier models – highly capable, but also very efficient, fast and at a lower price – is because we want to bring it to as many people as possible.”
