Google has officially started rolling out Gemini Spark to Google AI Ultra subscribers in the United States, marking one of the most ambitious consumer AI products the company has shipped to date. Announced at Google I/O 2026 on May 19, Spark is a personal AI agent that runs around the clock in Google Cloud — without requiring any device to stay powered on.
Key Highlights
- Gemini Spark operates 24/7 in Google Cloud, independent of your devices
- Integrates natively with Gmail, Docs, Sheets, Slides, Calendar, Drive, YouTube, and Maps
- Third-party app connections supported via the MCP protocol
- Android Halo overlay tracks agent activity on mobile
- Included in Google AI Ultra at $100/month — reduced from $250 at I/O 2026
How Gemini Spark Works
Unlike traditional AI assistants that respond only when prompted, Gemini Spark runs on dedicated virtual machines inside Google Cloud. This means it can act on your behalf even when your laptop is closed and your phone is in your pocket.
Built on Gemini base models and Google's internal Antigravity agentic harness, Spark can monitor inboxes, draft and send emails, schedule meetings, track deadlines, and execute multi-step workflows — all without requiring you to stay at the keyboard.
Users interact with Spark through the Gemini app or by emailing it directly at a dedicated address. On Android, the new Halo overlay interface lets you check what Spark is doing in the background without switching apps.
"It's your personal AI agent that helps you navigate your digital life, taking action on your behalf and under your direction," said Google CEO Sundar Pichai at I/O 2026.
Integrations and Developer Access
Spark connects natively with the full Google Workspace suite — Gmail, Docs, Sheets, Slides, Calendar, Drive, YouTube, and Google Maps. For developers, Spark also supports third-party service connections through the MCP protocol, signaling Google's intent to build an open agentic ecosystem.
Google has confirmed API access for enterprise customers and developers is in the pipeline, with broader availability outside the US expected in the coming months.
Privacy and Control
The shift toward always-on agents raises legitimate questions about privacy and oversight. Google has addressed this directly: all Spark activity is logged and visible in the Gemini app, and users can pause or cancel any task at any time. Explicit permission is required before Spark can access any new service or take irreversible actions.
Pricing and Availability
Gemini Spark is currently available to Google AI Ultra subscribers in the United States. At I/O 2026, Google cut the Ultra plan price from $250 to $100 per month to make the agent more accessible. The rollout began the week of May 26, 2026, and is expanding to more users throughout June.
Impact on the AI Landscape
The launch of Gemini Spark represents a fundamental shift in how AI assistants are designed — from tools you use to agents that work for you. With major competitors including Microsoft Copilot and Anthropic's Claude also investing heavily in agentic capabilities, the race for the personal AI agent category is now firmly underway.
For businesses and professionals, a 24/7 cloud agent that handles email, scheduling, and task execution autonomously could dramatically change how knowledge work gets done. The question is no longer whether AI can do these tasks — it is how much oversight people want to exercise when it does.
What's Next
Google plans to expand Gemini Spark availability to additional countries in the coming months, with broader third-party MCP integrations and additional Omni model variants on the roadmap. Enterprise-tier access with custom workflows and admin controls is also expected before year-end.
Source: Google Blog