Just ahead of Google's annual major I/O Developer Conference, a "big reshuffle" in AI usage rules has already quietly taken place. In addition to the highly anticipated upgrades to the Gemini large model and new hardware arrangements, Google has unexpectedly imposed a "usage limit": the Gemini that once allowed users to use freely is now saying goodbye to its "nearly unlimited" free period.

With the latest changes, the usage limit page for Gemini now features a more detailed counter. The new rules are no longer just about "chat number limits," but instead introduce a dynamic deduction mechanism. This includes two "5-hour cycle usage" and "weekly total quota" gates. This means that regardless of whether you're performing complex tasks or frequent interactions, the higher the computing power you consume, the faster you will reach the limit. Even if the counter resets within a short period, once the weekly limit is reached, you will still be locked out by the system.

Google's large model Gemini

Faced with this change, Google's intention is clear — it aims to guide high-intensity users to upgrade from the free tier to a paid subscription through the "Google AI Plus" plan. For users who do not want to pay, Gemini will have to step down from being the preferred AI tool to just one option among many.

This adjustment is not an isolated case, but rather part of a broader industry-wide "tightening" trend in AI. Recently, mainstream AI development environments like Windsurf have also started to reduce the benefits of their free tiers, from limiting concurrent usage to compressing context. Major manufacturers are all trying to filter core users by raising the bar.

This trend reflects a fundamental shift in the mindset of the tech industry: after spending billions on computing power and R&D over four years, investors' patience with AI is wearing thin. They are no longer satisfied with mere expansion, but are urgently seeking real profit prospects.