Google test new AI-based search mode

By Alimat Aliyeva

Google will begin testing a new search mode based on artificial intelligence, which allows users to ask questions consisting of multiple parts, Azernews reports.

The new feature, known as AI Mode, simultaneously runs several related search queries in the background, anticipating which subtopics users will be interested in next. It then presents a holistic answer that summarizes multiple search directions.

This new tool is aimed at competing with popular AI-based services like Perplexity and ChatGPT Search from OpenAI.

Robby Stein, Vice President of Google Search Products, highlighted that this feature draws on both the internet and the latest Google information systems. AI Mode can process text, images, and videos, and operates on Gemini 2.0, Google’s latest flagship artificial intelligence model.

According to the company, the new feature utilizes a special version of the Gemini 2.0 model, optimized for advanced reasoning and multimodal analysis. AI Mode can handle queries requiring in-depth analysis and comparison. The technology uses a "branching" method, executing several related search queries for different data sources at the same time and combining the results into a clear, concise answer. The system accesses not only web content but also Google's Knowledge Graph, real-world information, and data on billions of products.

As part of its future development, Google plans to make interactions with AI Mode more visual and expand its sources of information, including user-generated content. The system will also be refined to determine the optimal response format—ranging from hyperlinks for reservations to images and videos for visual instructions—based on the nature of the query.

This advancement could represent a significant leap in how users interact with search engines, moving beyond traditional keyword-based searches and offering more intuitive, personalized, and comprehensive answers. Google's AI Mode may transform the search experience, providing dynamic answers that cater to complex queries across multiple media types and improving the overall efficiency of online searches.

Google’s focus on multimodal responses could pave the way for even more immersive and interactive search experiences, potentially integrating real-time data, such as product availability or live events, directly into search results. This move signals a broader trend in the search industry, as companies increasingly rely on AI to enhance user experiences and improve the accuracy and depth of information retrieval.