The proactive AI assistant Gemini Spark, which Google unveiled at I/O 2026, has now officially entered the Apple ecosystem. Google announced that this “always-on, task-autonomous” AI agent has brought background automation capabilities to macOS, allowing Apple users to experience its powerful cross-device workflows.

Gemini Spark launches on macOS, priced at $100 per month
On Mac, Gemini Spark can directly access local files you authorize and complete multi-step tasks through natural language prompts. For example, you can ask it to create a budget spreadsheet using the latest invoice file on your computer, or organize messy folders by categorizing and sorting all Word documents in the Downloads folder—the entire process requires no manual operation, just one command. Google emphasizes that Spark only accesses files you explicitly authorize, and all sensitive information is processed locally to avoid unnecessary data leakage.

Google also revealed that the team is developing a new capability: allowing users to remotely control Spark via their phone, completing complex tasks without touching the computer. This means you can ask Spark to search for specific sales reports on your Mac, pull total revenue data, and send the organized information to your email—all while you’re away. This cross-device, cross-context automation makes AI assistants truly integrated into daily workflows, rather than just a tool confined to a single device.

Beyond local tasks, Spark has also begun deeply integrating with Google’s own services. It can read list contents from Google Keep and automatically create corresponding to-do items in Google Tasks, making the workflow between note-taking and task management much smoother. This cross-service automation makes Spark more than just executing commands—it can proactively coordinate information between different tools.

Third-party integrations are another key focus of this update. Spark already supports popular apps like Canva, Dropbox, Instacart, OpenTable, and Zillow Rentals. Through these integrations, you can ask Spark to design flyers in Canva, share files in Dropbox, make restaurant reservations on OpenTable, order groceries through Instacart, and even schedule home tours on Zillow Rentals. These features are currently available on web and mobile, with a Mac version coming in a future update.

Spark’s capabilities go beyond just executing tasks—it can actively monitor web information and respond to events in real time. According to Google, Spark can track a wide range of topics including sporting events, weather changes, financial markets, shopping updates, and blog posts, providing instant analysis when something happens without you needing to refresh your browser. For instance, when a stock crosses your set price threshold, Spark can instantly generate a financial summary; or immediately produce a post-game analysis report after a soccer match ends.

Currently, access to Gemini Spark remains limited: it’s only available to US users aged 18 and older with English as their primary language, and requires a subscription to the $100/month Google AI Ultra plan. Despite the high barrier to entry, Spark’s cross-platform capabilities and deep automation are gradually showcasing Google’s long-term vision for “proactive AI assistants”: not just having AI answer questions, but having it truly do the work for you.
Source: KOCPC Chinese