Google Adds Built-in Computer Use Tool to Gemini 3.5 Flash
*Google has introduced a computer-use capability directly inside Gemini 3.5 Flash.*
Google announced the feature on June 24. The update gives the model a built-in tool that lets it interact with a computer interface.
The change appears in both the Google DeepMind and Google corporate blogs. Hacker News placed the story on its front page the same day, where it drew 197 points and 128 comments.
No further technical details, benchmarks, or rollout timeline appear in the source materials. The posts describe the addition only as a “built-in computer use tool.”
Why it matters
The move places Google among the small group of labs shipping agent-style control of a desktop or browser from inside a shipping model. Engineers who already use Gemini 3.5 Flash will now test whether the new tool reduces the need for separate scripting or browser-automation layers. Until Google publishes usage data or safety evaluations, the practical scope of the feature remains unclear.
---
Sources:
{
"excerpt": "Google has added a built-in computer use tool to Gemini 3.5 Flash, allowing the model to interact directly with computer interfaces.",
"suggestedSection": "ai",
"suggestedTags": ["gemini", "google", "computer-use"],
"imagePrompt": "An abstract workspace with a glowing screen reflection on a dark desk surface and a single hand hovering near a keyboard. Muted color palette, cinematic lighting, 16:9."
}
No comments yet