Leadership1 publisher2 min readPublished
Google prices AI Ultra for developers at $100 a month a person
Google has published a price for frontier-tier access aimed at technical staff, and the figure a manager has to defend is per person per month, on a plan whose usage limits are quoted as multiples of a Pro allowance the company has not spelled out.
The Board Room · Leadership desk
What happened
- Google launched a $100-a-month AI Ultra plan at I/O 2026, available the same day, aimed at developers, technical leads, knowledge workers and advanced creators.
- The plan carries five times the Pro plan's usage limits in the Gemini app and Google Antigravity, alongside priority access to Antigravity and 20TB of cloud storage.
- Google cut its top AI Ultra plan from $250 to $200 a month with the capabilities unchanged, including twenty times the Pro plan's usage limits.
Compiled by The Board RoomSomething wrong?How this is made
Why it matters
- cost The commitment is recurring and scales with headcount, so the engineering budget carries it every month whether or not usage justifies the seat.
- constraint The allowance is sold as a multiple of Pro's limits, and Google keeps those numbers unpublished, so a buyer cannot size what it is purchasing before it purchases it.
- decision Someone now has to sort staff between the two tiers on evidence of who exhausts their allowance, and most teams lack that usage data.
- precedent Google has already repriced the top of its own ladder inside one product cycle, so a multi-year seat commitment should be negotiated on the assumption the tiers move again.
At $100 a month, one person costs $1,200 a year [10]. Forty of them is $48,000 a year, and the same forty on the $200 plan is $96,000 [11]. This is recurring spend, and the cheaper tier was available the day Google announced it [1], so an approval can land inside a quarter whose budget is already set.
The two tiers are priced out of proportion to what they allow. Google sells five times the Pro plan's usage limits for $100 and twenty times for $200 [2][5]. Divide the price by the multiple and the lower tier works out at $20 per unit of Pro-multiple against $10 at the higher one [12]. For the people who genuinely exhaust their allowance, the expensive plan is the cheaper one per unit of usage, and which staff those are is an empirical question most teams cannot yet answer.
Both figures are multiples of the Pro plan. The I/O post does not state Pro's limits in absolute terms [9].
The product is shaped like a consumer subscription. The $100 plan comes with 20TB of cloud storage and a YouTube Premium individual plan [4], Gemini Spark is listed as US only [6], and so is AI Inbox in Gmail [15]. That makes the price awkward for a technical leader. A published $100 means declining it is declining a known figure, and approving it means part of the line item is ad-free YouTube for staff.
For anyone modelling more than a year, the repricing matters more than the launch. The top tier fell from $250 to $200 and the capabilities stayed the same [5], a cut of 20% [13]. The ladder moved too: the $200 plan now buys four times the usage multiple of the $100 plan [14].
Availability is still in motion. Gemini Spark goes to trusted testers this week and is planned as a beta for US AI Ultra subscribers next week [6]. Project Genie is rolling out to eligible $200 subscribers aged 18 or over, globally [7]. Gemini 3.5 Flash sits in the $100 bundle for testing, debugging and iteration, and Google also lists it for AI Plus and Pro subscribers [8].
In my view the defensible move this quarter is a small number of $100 seats, for the staff who already hit Pro's ceiling. Hold everyone else at Pro until there is consumption data to argue from. Approve broadly now and the seat count is fixed by renewal, on a ladder that has been repriced once already [13].
What to watch
- Whether Google publishes Pro's absolute usage limits. Until it does, the 5X and 20X figures cannot be checked against measured consumption.
- Whether the $100 tier turns up in Workspace or enterprise agreements; the I/O post describes individual Google AI subscriptions.
- Gemini Spark's move beyond the planned U.S. beta. That move is what decides whether distributed teams can buy the same plan everywhere.