Skip to content

AI-enhanced Google Search now doubles as a personal digital assistant

With your consent, it is capable of placing calls to businesses for obtaining pricing and availability details.

Artificial Intelligence Integration in Google Search Transforms It into a Personal Assistant
Artificial Intelligence Integration in Google Search Transforms It into a Personal Assistant

AI-enhanced Google Search now doubles as a personal digital assistant

In an exciting development, Google has announced the rollout of an update for its Search and AI Mode service, effective today, July 15. This update significantly enhances the capabilities of the AI, making it more versatile and useful in a wide range of tasks.

One of the key improvements is in the area of math and coding. The integration of Google's most advanced AI model yet, Gemini 2.5 Pro, offers enhanced reasoning and stronger abilities in these areas. This upgrade provides a smarter alternative to the default search experience, particularly useful when dealing with complex questions or problems that require deeper analysis.

Another significant enhancement is the introduction of Deep Search. This tool, available primarily for users with Pro or Ultra subscriptions, is a powerful research tool that runs hundreds of mini-searches behind the scenes. It then organizes the information into a fully cited summary, ideal for tasks like planning major purchases or work projects.

In addition to these improvements, the update also focuses on business-related tasks. Google's AI now has the capability to call local businesses on users' behalf to gather information such as pricing and appointment availability. This feature is integrated into Search and aims to streamline tasks by automating interactions with businesses. Users are prompted with a questionnaire to customize the request before the AI makes the call.

The update also brings some changes to AI Mode. It is receiving an upgrade with the Gemini 2.5 Pro model, which improves reasoning for math and coding questions. Furthermore, the integrated AI model in AI Mode was added to AI Overviews, handling complex queries more efficiently.

For users with AI Pro or AI Ultra, there's an added benefit. They can swap to the 2.5 Pro option in AI Mode if they've enrolled in the Search Labs experiment. This upgrade also introduces higher limits for AI services for these subscribers.

In summary, these updates aim to make Google Search more functional as a personal assistant, capable of handling a variety of tasks beyond simple queries. The AI's "Shopping Graph" in AI Mode also browses for inspiration, thinks through considerations, and narrows down products during shopping, making online shopping more convenient.

[1] Google Search Blog: https://blog.google/products/search/google-search-gets-smarter-with-ai-updates/ [2] The Verge: https://www.theverge.com/2022/7/15/23266672/google-search-ai-mode-gemini-2-5-pro-update-features [3] TechCrunch: https://techcrunch.com/2022/07/15/google-search-ai-mode-update/ [4] CNET: https://www.cnet.com/tech/services-and-software/google-search-ai-mode-update-lets-the-ai-call-businesses-for-you/

  1. The integration of Google's advanced AI model, Gemini 2.5 Pro, not only improves the capabilities of the AI in math and coding but also expands its functionality in artificial-intelligence tasks, such as calling local businesses for pricing and appointment information.
  2. Deep Search, a powerful research tool available primarily for Pro or Ultra subscribers, uses artificial-intelligence to run hundreds of mini-searches behind the scenes, organizing the information into a fully cited summary, ideal for tasks like planning major purchases or work projects.

Read also:

    Latest

    Aligning Multiple Dimensions

    Alignment of Multidimensional Structures

    A manifold, in simpler terms, refers to a complex mathematical space that resembles an n-dimensional surface with certain constraints. It's the most general type of space where one can perform dimensionality reduction while maintaining correspondences across multiple datasets.