Google Enhances Chrome with AI Features

Google introduced three new AI-based features for the desktop version of Chrome: the Lens service for image analysis, the Tab Compare shopping assistant, and natural language integration for search history.

In the coming days, Lens will be available to users worldwide in the address bar and three-dot menu. The new tool offers enhanced functionality for searching information after selecting a part of the screen or clicking on an object:

• users can select part of a page and ask additional questions;
• there is an option to click on an object, such as a backpack in a picture, and find similar items in different colors from various brands;
• with Lens, users can learn additional information, such as how much sunlight a plant needs or how to solve a mathematical equation.

Previously, Google trained artificial intelligence to handle complex mathematical tasks.

Among other innovations is the Tab Compare feature to assist with shopping. Using artificial intelligence, Chrome will offer a summary of similar products from multiple tabs, including product characteristics, price, ratings, and features. All information will be compiled in one place as a comparison table.

The third feature is an improved search history using AI. If a user wants to return to a previously visited site but did not save it and forgot the name, they can use a query like: “What ice cream shop did I visit last week?” The browser will display the relevant pages.

The last two features will initially be available in the United States.

Experimental Gemini 1.5 Pro Surpasses GPT-4o

Google DeepMind’s AI division has released an experimental version of its flagship AI model, Gemini 1.5 Pro.

We are increasing the rate limits for the new Gemini 1.5 Pro model in Google AI Studio. Keep the feedback coming ↓ https://t.co/WsaLFmynrC

— Google DeepMind (@GoogleDeepMind) August 1, 2024

The new version of the neural network has attracted public attention due to its high test results. Developer Logan Kilpatrick noted that Gemini 1.5 Pro ranked first in the LMSYS rating, which tests AI on various tasks and assigns an overall score.

It is unknown whether the experimental version of the neural network will be used by default. It is in the testing phase and may be withdrawn or altered.

Previously, DeepMind’s robotics division implemented Gemini 1.5 Pro in a robot to teach it to navigate an office and respond to commands.

In August, Google released three new open-source generative AI models.

In June, the corporation added the Gemini AI model to the sidebar of Gmail, Docs, Sheets, Slides, and Drive for paid users.