According to a report by Android Authority, Google is preparing to make a big change for how Google works with its Gemini AI assistant Android smartphone. From 7 July, Gemini will be able to interact directly with core apps such as phones, messages, WhatsApp and other utility apps, even if a setting called “Gemini Apps Activity” is on or closed. This update has raised questions among users, especially around privacy and data control.

The news came to light after some users received an email from Google’s Gemini team. The message mentions that Gemini will help you use some apps from July 7. However, email clearly did not explain what “helping you” means. For example, the users are uncertain whether it means that Gemini will be able to read or send messages, reach the contacts, check the call log or control other personal features on the phone.

Another thing of confusion is how to close it. While email says that users can disable these features with the “apps settings page”, it does not at all explain where this page is located or which specific option needs to be closed. This lack of clarity has surprised users, especially since the changes will allegedly apply, even though someone has already disabled the Gemini app activity.

The email also included a link to Google’s Gemini Apps Privacy Hub. But the users going on the page say that it does not actually provide a lot of clarity on what it is changing or at what level of Gemini will reach. Many people are wondering whether Mithun will now be able to manage interactions in WhatsApp or control calls and messages even without full user permission.

With updates around the corner, many users are asking Google to be more transparent and provide a detailed description. Tech tips like CID on X (East) have flagged off the issue, which regularly adds growing anxiety among Android users using Gemini. Some people say that they have not received email at all, which adds more confusion about on whom this change applies and how wider the update will be.

With July 7, Android users are advised to check their device settings and be cautious for new updates, especially concerned about privacy and app permissions. Whether the new features of Gemini become helpful or the possibility of infiltration will depend on how the transparent and user -friendly Google makes the rollout.

In addition, Google recently made a major announcement at robotics space. Its research branch, Deepmind introduced a new version of Gemini, called Gemini robotics on-device. Unlike the previous AI model, which required internet access, this version can walk directly on the robot without the need to connect online.

This local version makes the first Gemini robotics model launched in March and is designed to handle tasks through the natural voice command. Developers can train models to do tasks such as clothes or unknown bags only using spoken instructions. Google was also being used by AI on various types of robots-from Aloha robot to B-Arme Franca FR3 and Apollo Humanoid Robot of Aptronic.

The company claims that the on-device model is almost better than its cloud-based version, as well as, and in some cases, even better than other local AI models. Although Google has not named which other models compared it, highlighted that their model could be suited to new tasks, even those people have not faced it.

In one case, the B-arm robot was able to complete the assembly-line functions, which were not specially trained. This suggests that Google is not only focusing on AI that can help smartphone users, but can also focus on robots that may suit real -world functions on the fly.