Google Brings Computer Use to Gemini on macOS: Agentic AI Executes Multi-Step Tasks
⚡ Breaking News
TechCrunch AI
November 20, 20264 min read0

Google Brings Computer Use to Gemini on macOS: Agentic AI Executes Multi-Step Tasks

By — Senior AI Technical Editor

Back to News
❝

Google launched Computer Use in Gemini for macOS on Thursday, enabling the AI to perform multi-step tasks across native apps and websites in the background. The feature is initially available in English in the US only, requiring a Google AI Plus, Pro, or Ultra subscription. It arrives later than similar agentic features from OpenAI, Anthropic, and Meta.

Executive Overview

Google officially announced on Thursday the launch of Computer Use in the Gemini app for macOS, enabling the AI to execute multi-step tasks directly on the user's device. The feature is initially available in English in the United States only and requires the Gemini app for Mac with a Google AI Plus, Pro, or Ultra subscription. This launch positions Google as a relative latecomer compared to OpenAI, Anthropic, and Meta, which have offered similar agentic features in desktop apps for some time.

📊 Official Technical Specifications & Data Sheet

Technical AspectConfirmed Official Data
💰 Pricing & Usage CostAvailable only with Google AI Plus, Google AI Pro, and Google AI Ultra subscriptions; not available on the free plan per the announcement. No additional token cost is imposed for this feature within the subscription.
🌐 Platforms & Immediate AvailabilityGemini app for macOS only. Initial launch: English + United States. No announcement for Windows, Linux, or web.
⚡ Performance & Speed BenchmarksNo official benchmark figures published in the announcement; the feature operates with a live preview inside the chat, allowing screen and cursor editing to continue work.
🛡️ Security & ControlAbility to pause, fully stop, and pre-approve each action before execution, similar to other computer agents.
🧠 Context WindowThe context window size for this feature was not disclosed in the source report.
🌍 Arabic Language & Regional SupportNot currently supported; launch is limited to English in the United States with no timeline for regional expansion.

Deep-Dive Features & Architecture

The Computer Use feature allows the Gemini agent to work directly on the Mac by navigating native applications and browsing websites in the background, enabling the user to continue working on the screen and cursor while the agent executes the task in a live preview window within the chat. Google's suggested use cases include filling out forms, repeating the same tasks across batches of files, organizing files into folders according to specific criteria, and creating presentations from files or online dashboard data.

This launch follows Google's announcement this week of the broader expansion of the Tasks feature within the app, which allows the AI to log into websites, make purchases, and book appointments. Google also recently began accepting early testers for the CC AI agent announced in September to help families organize across email, calendar, chats, and tasks. This launch places Google in a relatively late position compared to OpenAI, Anthropic, and Meta, which have offered similar agentic features in desktop apps for some time.

Benchmark & Competitive Performance

According to the source report, Google is considered late in this field compared to OpenAI, Anthropic, and Meta, which have provided agentic features in desktop apps for a while. On the other hand, Google excels by integrating the feature into its existing subscription ecosystem (AI Plus, Pro, and Ultra) without announced additional token costs, with explicit control measures (pause/approve) that enhance transparency. The announcement does not include standard benchmark numbers for comparison, making competitive evaluation qualitative so far.

Industry Impact & Enterprise Adoption

The introduction of Computer Use in Gemini for macOS signals Google's push into the agentic AI space, where AI agents autonomously perform tasks on behalf of users. For enterprises, this could streamline workflows such as data entry, file management, and report generation, reducing manual effort. However, the initial limited availability (English, US only) and subscription requirements may slow widespread adoption. The feature's security controls, including pause and pre-approval, are crucial for enterprise environments where compliance and oversight are paramount. As competition intensifies, Google's integration with its subscription services may offer a cost advantage, but the lack of benchmark data leaves performance comparisons uncertain. Enterprises should monitor expansion to other languages and regions, as well as future benchmark disclosures, before integrating into critical workflows.

Conclusion

Google's move toward computer agents on macOS comes competitively late but is structured with clear controls and subscription integration. While the feature is currently limited to English in the US, it represents a significant step in Google's agentic AI strategy. The absence of official benchmarks and Arabic support limits immediate global appeal, but the potential for workflow automation is substantial. As Google expands availability and refines the feature, it could become a key differentiator in the AI assistant market. For now, users in the US with eligible subscriptions can experiment with Computer Use, while others await further announcements.

Media Source: TechCrunch AI | Fact Verification & Analysis: AI Tools Oasis

Original Source:TechCrunch AIThis news was formulated based on coverage from TechCrunch AI

Frequently Asked Questions

What is the new Computer Use feature in Gemini on macOS?

It is an agentic feature that allows Gemini to execute multi-step tasks directly on a Mac by navigating native apps and browsing websites in the background, with a live preview inside the chat and the ability to pause or approve each action before execution.

Which Google subscriptions are required to use the feature?

The feature requires the Gemini app for Mac with a Google AI Plus, Google AI Pro, or Google AI Ultra subscription. It is not available on the free plan according to the current announcement.

Where and when is Computer Use in Gemini available?

It is initially available in English in the United States only, as of Google's announcement on Thursday. Google has not specified a timeline for expanding to other markets or platforms.

Does the feature support Arabic or the Middle East region?

No. The initial launch is limited to English in the United States, and Google has not announced Arabic support or regional availability in the Middle East yet.

How does Computer Use differ from the Tasks feature in Gemini?

Tasks works within the app to log into websites, make purchases, or book appointments, while Computer Use actually controls the Mac to navigate local apps, files, and the web, enabling broader workflow automation such as form filling, file organization, and presentation creation.

AI Tools Oasis

AI Tools Oasis Team

Bringing you the latest news and analysis in the world of Artificial Intelligence with accuracy and credibility. Follow us for all updates.

Related News