Google's AI Mode gets agentic capabilities in India

Earlier this week, Google integrated the latest generative Artificial Intelligence (gen AI) Large Language Model (LLM)-based Gemini and the propreitary Google Graph with the AI mode and Circle to Search features to improve the shopping experience online.

Now, Google has announced to introduce new agentic capabilities to AI Mode to enhance the user experience on the search app.

For the uninitiated, agentic AI is next-generation bot, which can perform tasks withs minimal supervision. Once it has been asked to complete a job like adding a new feature to an app, it is capable of understanding every aspect of the work, and even if it faces any issues, the AI is capable of finding a solution, improvising the code and completing tasks in less time.

In this case, the new AI Mode can help regular users find a good restaurant.

The user can ask about getting a dinner reservation with friends that includes multiple constraints and preferences — like party size, date, time, location and preferred cuisine — and AI Mode will streamline this process.

It will search across multiple reservation platforms and websites, and it will find real-time availability for restaurants that meet the user's specific needs and then present it with a curated list of restaurants with available reservation slots to choose from.

The AI Mode does the legwork and links the user directly to the booking page, so he/she can easily take the last step and finalise the reservation.

Under the hood, AI Mode uses the live web browsing capabilities of Project Mariner, direct partner integrations on Search. It also makes good use of the company's propreitary Knowledge Graph and Google Maps to help users take action on the web. In India, Google is working with partners like EazyDiner, Swiggy, and Zomato to make an agentic AI experience possible.

In a related development, Google has launched Gemma AI-powered AI Edge Eloquent, a speech-to-text dictation app for iPhones.

It is free to download the app, and it can work offline. Language processing happens on-device, so the user's audio and the generated text are available on iPhone at all times. However, for advanced optional features, it may be relayed to the server.

Regular dictation applications usually stumble and fail to provide filler words for 'ums', 'uhs', and other types of pauses. Thanks to Gemma 4's advanced Automatic Speech Recognition (ASR) capabilities, it can understand the user's intended meaning and edit out those mid pauses, polish it with the right word in the speech or the interview.

Get the latest news on new launches, gadget reviews, apps, cybersecurity, and more on personal technology only on DH Tech.