Google Gemini's New Quota Model Sparks Backlash Over Transparency and Restrictiveness

May 26, 2026
Google Gemini's New Quota Model Sparks Backlash Over Transparency and Restrictiveness
  • The shift has fueled trust concerns, as Google framed it as cost-driven but communicated the change via quiet emails and vague support pages, reducing perceived predictability for subscribers.

  • Previously, Gemini raised quotas for select Antigravity users, but regular AI Pro subscribers did not receive a broad update.

  • The core debate centers on predictability and transparency of the new quota system, with users struggling to estimate how much quota a given task will consume.

  • The rollout of Google Gemini’s new compute-based quota model has sparked user backlash over perceived restrictiveness and a lack of clarity in how limits are calculated, prompting questions about transparency and pricing.

  • Paid tiers significantly expand compute access: AI Plus roughly doubles standard limits, AI Pro about quadruples, and AI Ultra ranges from 5x to 20x Pro limits, with a new product-based credit model for tools like Flow and Antigravity that can grant extra credits after caps.

  • Users on forums report limits being exhausted quickly, sometimes after a few advanced tasks such as PDF creation, image generation, or coding help, and point to failed responses or server errors still counting toward the quota.

  • New account storage trials show some users not linked to phone numbers may see 5GB of free storage instead of the usual 15GB across Google services.

  • A user-shared video demonstrating the avatar feature draining quota rapidly drew immediate attention from Google staff.

  • The controversy raises broader questions about transparency, pricing, and fair usage in AI subscription services as competitors implement similar usage windows affecting long conversations and professional workflows.

  • Reports show mixed experiences: some Antigravity subscribers enjoy higher quotas while regular subscribers await clearer rules and updates.

  • Google has not issued a detailed public clarification on how intensive generation tasks, like avatar-based video creation, impact usage limits under the AI Pro plan.

  • There are reports of unintended model switching, with paying for Pro access sometimes falling back to a lighter Flash model without notice, adding to user dissatisfaction.

Summary based on 5 sources


Get a daily email with more Tech stories

More Stories