Google has begun opening Gemini Spark to subscribers on its Google AI Pro plan in the United States, ending a stretch in which the agentic assistant sat behind the company's most expensive subscription. Word of the change came directly from Google's official Gemini account on X, which confirmed the US Pro rollout and indicated that Pro subscribers in additional countries would gain access before long. For anyone paying $20 a month rather than stepping up to the Ultra tier, it's the first practical chance to see whether the feature Google showcased at I/O in May holds up in daily use.
What Gemini Spark Actually Does Differently
The standard Gemini chatbot responds to prompts. Spark is built around a different premise: rather than returning an answer and waiting for the next question, it carries out work on your behalf. Hand it an assignment — clearing out an overloaded inbox, making changes to your calendar — and it proceeds with the job independently.
Background execution without an open app
One of the more consequential design decisions here is that Spark keeps running after you've closed the app. The task doesn't stall because you switched contexts or put the device down. That persistence is what separates an agent from a chat window, and it's the reason the feature has drawn as much attention as it has since its debut.
Trigger-based automation
Spark can also be pointed at conditions rather than commands. Configure it to monitor for a particular event — a delayed flight being the example Google leans on — and it responds on its own when that event occurs. The user sets the rule once; the agent handles the watching and the reacting.
The Gemini Spark Rollout Timeline
The path from announcement to broad availability has been deliberate, with Google widening the circle in stages rather than all at once:
|
Stage
|
What happened
|
|
I/O in May
|
Spark introduced to the public
|
|
Later in May
|
Released to Google AI Ultra subscribers in the US
|
|
June
|
Platform support extended to macOS
|
|
Earlier this month
|
Reached Ultra subscribers in additional regions; language support broadened
|
|
This week
|
First rollout beyond the Ultra tier, to US Pro subscribers
|
That final row is the meaningful one. Every prior expansion moved Spark across platforms or borders while keeping it locked to Ultra. The Pro release is the first time the price barrier itself has come down, and Google is expected to keep extending regional coverage over the coming weeks.
Where Gemini Spark Still Isn't Available
Wider access hasn't translated into universal access. Anyone in the European Economic Area, Nigeria, Switzerland, or the United Kingdom remains shut out of Spark entirely — and no amount of subscription spending changes that. The restriction applies across Google AI plans, so upgrading from Pro to Ultra does nothing for a user in London or Zurich.
Google has offered no timetable for lifting these regional exclusions, and has stopped short of committing that they will be lifted at all. For users in those markets, the practical takeaway is that Spark's status is unresolved rather than merely delayed.
What the Pro Tier Move Signals About Agentic AI
Pushing Spark down to the $20 plan reads as a statement of intent. Agentic capabilities have generally been positioned as premium differentiators, reserved for the subscribers paying the most, and keeping Spark on Ultra fit that pattern neatly. Moving it to Pro breaks it — an indication that Google would rather build habitual use across a large base than preserve the feature as a flagship exclusive.
For US Pro subscribers, the recommendation is simply to try it. Spark arrived with considerable expectation attached, and the only way to judge whether background task execution and trigger-based automation deliver on that expectation is to put real work through it. Everyone else on Pro outside the US is looking at a wait measured in weeks rather than months, based on the expansion pace Google has signaled.

