Generate data with Gemini in Google Sheets

What’s changing

Currently, you can utilize Gemini in Google Sheets to generate charts and valuable insights, and today we’re excited to bring AI directly into the cells of your spreadsheet with the power of AI function. 

From a cell, you can now input a specific prompt describing what you would like to do and an optional data range, and Gemini can do the following: 
  • Use relevant information from your spreadsheet to generate text that is tailored to your specific data. 
  • Summarize information by analyzing the content of your spreadsheet and highlighting the most important parts, so you can quickly understand your data. 
  • Categorize your data into meaningful groups, so you can easily detect patterns, such as user sentiment. ai functions in sheets

Here are some examples of how you’d use this feature: 
  • Text generation: 
    • =AI(“Write a formal ad copy for the product. Cater copy to the objective and target audience”, A2:C2) 
    • =AI("Create an email to the reviewer addressing specific items in their reviews.", A2:G2) 
  • Summarization: 
    • =AI("For the customer, write a one sentence summary of their feedback.", A2:D2) 
    • =AI("You are the owner of a pet sitting business. Write a 2 sentence summary for the customer about their pet’s last stay. Be a little funny.", F2) 
  • Categorization: 
    • =AI("Categorize the customer inquiry as a compliment, exchange request, or return request.",C2) 
    • =AI("Perform sentiment analysis on the emails sent by the customers to the barbershop.", C2)
If you want to refer to two ranges that aren’t contiguous, you can use concatenation in the prompt. Here are a few examples: 
  • =AI("Write a review of the hot sauce "&A2&" using only the ingredients "&D2&" and the popularity "&E2&".", A2:D2) 
  • =AI("Find the major themes in the customer feedback of "&B2&" using the rating "&C2&" and comments: "&D2&".") 

Who’s impacted 

End users 

Why you’d use it 

Now you can create hundreds of rows of data quickly to do things like categorize user feedback requests or write personalized emails, all without being an expert in data or spreadsheets. Regardless of your data needs, with instant access to Gemini from any cell in your Sheet, the Al function helps you get work done faster. 

Additional details 

  • We currently support text-based responses for text generation, summarization, categorization, and sentiment analysis queries. While the input information is optional, we recommend using it for optimal results. 
  • After generating, if an explicitly referenced cell or the prompt is updated, the cell shows an out-of-sync state, and keeps the previous output for reference. 
  • To refresh a cell, you can select one or multiple cells and press “refresh and insert.” 
  • You can select multiple cells with AI functions and generate outputs; however, only the first 200 selected cells with AI functions will be generated. You can wait for the generation to complete and select more cells to generate outputs. 

Getting started 

Rollout pace 

  • Rapid Release domains: Extended rollout (potentially longer than 15 days for feature visibility) starting on June 25, 2025 
  • Scheduled Release domains: Gradual rollout (up to 15 days for feature visibility) starting on July 29, 2025 

Availability 

Available for Google Workspace: 
  • Business Standard and Plus 
  • Enterprise Standard and Plus 
  • Customers with the Gemini Education or Gemini Education Premium add-on 
  • Google AI Pro and Ultra 
Anyone who previously purchased these add-ons will also receive this feature: 
  • Gemini Business* 
  • Gemini Enterprise* 
*As of January 15, 2025, we’re no longer offering the Gemini Business and Gemini Enterprise add-ons for sale. Please refer to this announcement for more details. 

Resources

Chrome Beta for Desktop Update

The Chrome team is excited to announce the promotion of Chrome 139 to the Beta channel for Windows, Mac and Linux. Chrome 139.0.7258.5 contains our usual under-the-hood performance and stability tweaks, but there are also some cool new features to explore - please head to the Chromium blog to learn more!

A partial list of changes is available in the Git log. Interested in switching release channels? Find out how. If you find a new issue, please let us know by filing a bug. The community help forum is also a great place to reach out for help or learn about common issues.

Chrome Release Team
Google Chrome

How Mecha BREAK is driving PC-only growth on Google Play Games

Posted by Kosuke Suzuki – Director, Games on Google Play

On July 1, Amazing Seasun Games is set to unveil its highly anticipated action shooting game - Mecha BREAK, with a multiplatform launch across PC and Console. A key to their PC growth strategy is Google Play Games on PC, enabling the team to build excitement with a pre-registration campaign, maximize revenue with PC earnback, and ensure a secure, top-tier experience on PC.

Building momentum with pre-registration

With a legacy of creating high-quality games since 1995, Amazing Seasun Games has already seen Mecha BREAK attract over 3.5 million players during the last beta test. To build on this momentum, the studio is bringing their game to Google Play Games on PC to open pre-registration and connect with its massive player audience.

"We were excited to launch on Google Play Games on PC. We want to make sure all players can enjoy the Mecha BREAK experience worldwide."
- Kris Kwok, Executive Producer of Mecha BREAK and CEO of Amazing Seasun Games
screenshot of Mecha BREAK's pre-registration on Google Play Games on PC homepage
Mecha BREAK pre-registration on Google Play Games on PC homepage

Accelerating growth with the Native PC program

Mecha BREAK's launch strategy includes leveraging the native PC earnback, a program that gives native PC developers the opportunity to unlock up to 15% in additional earnback.

Beyond earnback, the program offers comprehensive support for PC game development, distribution, and growth. Developers can manage PC builds in Play Console, simplifying the process of packaging PC versions, configuring releases, and managing store listings. Now, you can also view PC-specific sales reports, providing a more precise analysis of your game's financial performance.

Delivering a secure and high quality PC experience

Mecha BREAK is designed to deliver an intense and high-fidelity experience on PC. Built on a cutting-edge, proprietary 3D engine, the game offers players three unique modes of fast-paced combat on land and in the air.

    • Diverse combat styles: Engage in six-on-six hero battles, three-on-three matches, or the unique PvPvE extraction mode "Mashmak".
    • Free customization options: Create personalized characters with a vast array of colors, patterns and gameplay styles, from close-quarters brawlers to long-range tactical units.
moving image of Mecha BREAK's gameplay on Google Play Games on PC
Mecha BREAK offers a high-fidelity experience on PC

The decision to integrate with Google Play Games on PC was driven by the platform's robust security infrastructure, including tools such as Play Integrity API, supporting large-scale global games like Mecha BREAK.

"Mecha BREAK’s multiplayer setting made Google Play Games a strong choice, as we expect exceptional operational stability and performance. The platform also offers advanced malware protection and anti-cheat capabilities."
- Kris Kwok, Executive Producer of Mecha BREAK and CEO of Amazing Seasun Games

Bring your game to Google Play Games on PC

This year, the native PC program is open to all PC games, including PC-only titles. If you're ready to expand your game's reach and accelerate its growth, learn more about the eligibility requirements and how to join the program today.


Google Ads Scripts Sunsetting Ad Customizers for Expanded Text Ads and Legacy Ad Extension Entities Starting July 14, 2025

Starting July 14, 2025, Google Ads scripts will begin to throw errors for requests related to ad customizers for expanded text ads and feed-based legacy ad extensions.

Ad customizer changes

Any requests related to Ad Customizer entities, such as AdsApp.AdCustomizerSource or AdsApp.newAdCustomizerSourceBuilder, will begin to return sunset errors. These entities have been read-only since deprecation in 2022 and haven’t been served on any ads since May 2024. They were also fully removed from the Google Ads API in the v19 release.

Ad extension changes

Feed-based Ad Extensions were sunset in 2022 in favor of asset-based extensions. Users were asked to migrate, or were migrated automatically, but were able to access legacy functionality through the withOnlyLegacy() methods on various selectors. The withOnlyUpgraded() method was added to allow users to only use upgraded asset-based extensions. Starting July 14, 2025, the legacy functionality will be removed, and withOnlyLegacy() methods will begin to throw a sunset error. The withOnlyUpgraded() methods will log a warning, since it will no longer be necessary to specify the use of asset-based extensions when they are the only option.

If you have any questions, ask them in the Google Ads scripts forum. We're here to help!

 - Dora Sun - Google Ads API Team

Using KerasHub for easy end-to-end machine learning workflows with Hugging Face

KerasHub enables users to mix and match model architectures and weights across different machine learning frameworks, allowing checkpoints from sources like Hugging Face Hub (including those created with PyTorch) to be loaded into Keras models for use with JAX, PyTorch, or TensorFlow. This flexibility means you can leverage a vast array of community fine-tuned models while maintaining full control over your chosen backend framework.

Using KerasHub for easy end-to-end machine learning workflows with Hugging Face

KerasHub enables users to mix and match model architectures and weights across different machine learning frameworks, allowing checkpoints from sources like Hugging Face Hub (including those created with PyTorch) to be loaded into Keras models for use with JAX, PyTorch, or TensorFlow. This flexibility means you can leverage a vast array of community fine-tuned models while maintaining full control over your chosen backend framework.