Google is rolling out Gemini Labs, a dedicated home for experiments inside the Gemini web app, giving users a clear place to try early features and a new control for how personal data informs chats. The Labs area sits in the Tools menu with a beaker icon and ships alongside a Personal Intelligence toggle that lets you decide, per conversation, whether Gemini can draw from connected Google apps for context.
A Clear Home for Experiments Inside the Gemini Web App
Rather than blending test features with stable tools, Gemini Labs isolates work-in-progress capabilities in one labeled section. As spotted by 9to5Google, early entries include items like Agent and Dynamic View, signaling that Google wants testers to know exactly what’s experimental and what’s production-ready.
The Tools menu itself has been tidied up to make room for Labs. Core utilities now live in a standard Tools area, while Labs stands apart for opt-in experiments. This mirrors long-running programs such as Google Labs and Search Labs, which helped the company validate concepts before broader releases.
Personal Intelligence Puts Users In Control
Inside Gemini Labs, a Personal Intelligence toggle lets you choose whether Gemini can personalize answers by pulling context from connected apps like Gmail, Calendar, or Drive. It applies only to the current chat and is enabled by default; turn it off, and Gemini will stop referencing connected app data for that conversation. Start a new chat and the setting resets, reducing the chance of unintentionally carrying personalization forward.
Consider a travel planning example: with the toggle on, Gemini could surface your upcoming flight details from Gmail or suggest meeting times from Calendar. Turn it off, and the assistant sticks to general web or user-provided context. This per-thread approach aligns with privacy-by-design principles advocated by groups such as the Electronic Frontier Foundation and helps ensure transparency about when personal signals are used.
Why a Dedicated Labs Section Matters for Gemini Users
Separating experiments from stable features reduces cognitive load and clarifies expectations. Users know exactly where to test new ideas and what to trust for daily workflows, while Google can iterate faster without muddying core functionality. It also reflects a wider industry pattern: OpenAI exposes beta ChatGPT features behind explicit toggles, and Microsoft offers learning and trial spaces through Copilot Lab, both designed to gather feedback without destabilizing the main product.
There’s a regulatory and trust dimension too. Clear labeling, reversible controls, and session-scoped personalization help address concerns raised by consumer protection bodies about dark patterns and opaque data use. If you can see what’s experimental and decide, at the chat level, how much context Gemini gets, you’re more likely to test new tools confidently.
Availability Today and What Gemini Users Can Expect Next
For now, Gemini Labs and the Personal Intelligence toggle appear limited to the Gemini web app in select regions, with mobile clients unchanged. That staged rollout is typical for Google’s AI features and lines up with months of behind-the-scenes hints in app teardowns suggesting a centralized venue for opt-in experiments, including advanced capabilities for Gemini Live.
Expect Labs to expand as features mature and move into general availability. If the trajectory of Search Labs is any guide, Google will likely use this channel to pilot deeper agentic workflows, richer multimodal views, and integrations that benefit from real-world testing at scale.
The Bigger Picture for AI Assistants and Responsible Rollouts
Industry analysts forecast rapid enterprise uptake of generative AI, with Gartner projecting that by 2026 more than 80% of enterprises will have used GenAI APIs and models or deployed GenAI-enabled applications. In that context, a well-marked Labs lane is not just a UI nicety; it’s a product strategy for shipping responsibly while demand and capabilities accelerate.
Gemini Labs formalizes the testbed users have wanted: a place to experiment boldly without sacrificing clarity or control. With a per-chat Personal Intelligence switch and cleaner separation from stable tools, Google is signaling that speed and transparency can coexist in the next wave of AI assistants.