Google Brings Computer Management Right into Gemini 3.5 Flash
25.06.26
Google has announced a major update to its artificial intelligence ecosystem, adding an advanced computer management tool to Gemini 3.5 Flash. This solution replaced the experimental Gemini 2.5 framework, offering developers a more advanced visual approach to performing complex computing tasks.
What is the new computer management feature in Gemini 3.5 Flash?
The new functionality allows AI agents to act as “digital users.” Thanks to computer vision technology based on the analysis of screenshots, the neural network is able to see the active application window, understand the structure of interface elements and perform manipulations: click the mouse, scroll pages or enter text. This is full automation of actions that previously required manual control.
Main capabilities and applications of AI agents
The tool provides broad prospects for increasing efficiency in work processes. Key use cases include:
- Automatically filling out complex forms and transferring data between systems.
- Interaction with desktop and browser software to collect information.
- Software testing through visual control.
- Bulk data processing in interfaces without API.
Safety and protection against prompt injections
Giving control of the keyboard and mouse to an algorithm raises cybersecurity concerns. To avoid unauthorized actions (for example, clicking on dangerous links or activating hidden commands on malicious resources), Google has implemented multi-level protection:
- Corporate control: any significant or irrevocable transaction requires mandatory human confirmation.
- Attack monitoring: The system is equipped with prompt injection detection mechanisms that automatically block suspicious activity when attempting to hack the model.
Availability for developers
Currently the features are available through the Gemini API. For those wishing to test the new product, Google has created a demo platform based on Browserbase, where the functionality of Gemini Enterprise Agent can be tested in real time. This greatly simplifies the barrier to entry for developers looking to incorporate intelligent agent capabilities into their products.
Don't miss interesting news
Subscribe to our channels and read announcements of high-tech news, tes
Oppo A6 Pro smartphone review: ambitious
0 
Creating new mid-range smartphones is no easy task. Manufacturers have to balance performance, camera capabilities, displays, and the overall cost impact of each component. How the new Oppo A6 Pro balances these factors is discussed in our review.
In-Ear headphones Active Noise Cancellation: how Oppo eliminate noise
0 
In-ear headphones remain one of the most popular form factors due to their lightweight design. However, this design is considered one of the most challenging to implement active noise cancellation (ANC). We’ll explain how Oppo solved this problem.
Modular Fairphone 6+ Surfaces: Specs, Price, and Launch Date
Key details regarding the modular Fairphone 6+ from the Dutch manufacturer have leaked. The new model will feature an upgraded processor, increased RAM, and an exclusive color option.
NEC Launches First All-AI Department With 17 Autonomous Agents
NEC has launched an internal department of 17 autonomous AI agents, including a digital manager.