Google has rolled out major upgrades to Chat in NotebookLM, transforming it into a more powerful, goal-driven AI research partner. The enhancements, powered by the latest Gemini models, include an 8x larger context window, a 6x increase in conversation memory, and a 50% boost in response quality—especially when drawing from extensive source materials. The most significant update is the full integration of Gemini’s 1 million token context window across all NotebookLM plans. This allows the tool to process and analyze vast document collections with greater depth and accuracy. Users can now engage in longer, more coherent conversations without losing context, making it ideal for complex research projects. The system now automatically explores sources from multiple angles, uncovering deeper insights and connections beyond the initial query. This multi-perspective analysis helps users synthesize nuanced, well-rounded responses—particularly valuable when working with large or intricate datasets. Conversation history is now saved automatically, enabling users to pick up where they left off, even after closing a session. This feature supports long-term projects and iterative research. Users retain full control, with the ability to delete chat history at any time. In shared notebooks, chat remains private to the individual user. A new feature now available to all users allows setting custom goals, voices, or roles for each conversation. By clicking the configuration icon, users can define how they want the AI to behave. Examples include: Acting as a PhD research advisor who rigorously challenges assumptions and demands strong evidence. Taking on the role of a lead marketing strategist to deliver immediate, actionable plans with clear next steps. Analyzing content from three distinct viewpoints: academic, creative, and skeptical—ensuring thorough evaluation. Serving as a Game Master for a text-based simulation, crafting immersive scenarios with defined goals and constraints. These goal-driven interactions make NotebookLM more adaptable to diverse workflows, whether for academic research, strategic planning, or creative exploration. With these advancements, Google aims to make NotebookLM not just a tool for summarizing information, but a proactive collaborator that evolves with the user’s needs. The updates are rolling out gradually over the coming week, bringing a new level of intelligence, personalization, and productivity to AI-assisted research.