NotebookLM en 2026, lo que realmente merece la pena
This video explores NotebookLM 2.0's powerful features for 2026, demonstrating how to use curated sources to generate reports, presentations, Excel files, and vertical videos. The tool works best as a structured context-based system rather than a general chatbot, transforming sources into actionable insights and multiple formats while requiring critical review of outputs.
Summary
The transcript provides a comprehensive guide to NotebookLM's architecture and capabilities. The tool is structured in three parts: sources on the left (PDFs, videos, documents, links), a chat interface in the center that references specific source fragments through citations, and a studio on the right that converts information into reports, presentations, mind maps, infographics, videos, and new vertical shorts. The speaker emphasizes that NotebookLM differs fundamentally from general chatbots because it grounds responses in provided sources with specific citations. Featured notebooks from publishers and researchers offer pre-curated sources on trending topics, allowing users to ask practical questions grounded in real data rather than generic AI responses. Key practical applications include generating downloadable PDFs and PowerPoint presentations from sources, creating Excel files with editable financial models (demonstrated through a car purchase comparison), and producing visual charts from analysis. The speaker introduces three critical validation questions users should ask before trusting a notebook: detecting contradictions between sources, identifying gaps in coverage, and finding missing perspectives or alternative viewpoints. A powerful use case involves combining Gemini and NotebookLM to turn a creator's content into a private mentor—gathering recent videos from a channel, analyzing patterns in the creator's thinking, transforming those patterns into actionable principles, and converting the knowledge into various formats like infographics and presentations. The video also discusses Google's announced short video summary feature, converting complex sources into 60-second vertical videos for mobile learning. The speaker cautions that NotebookLM should not be used as a general chatbot but rather as a tool for working within closed contexts and specific sources. Importantly, outputs require verification—calculations, health claims, financial decisions, and client work must have all data, sources, and conclusions reviewed. The correct approach is using NotebookLM to eliminate slow initial work, organize information, detect patterns, and produce polished first versions rather than expecting final, perfect outputs.
Key Insights
- NotebookLM's value proposition differs fundamentally from general chatbots because it grounds all responses in specific source citations—users can click citation numbers to jump to exact fragments where information originated
- The speaker identifies three critical validation questions to assess source quality before trusting a notebook: detecting contradictions between sources, identifying coverage gaps, and finding missing perspectives or alternative viewpoints
- Excel file generation can create editable financial models with changeable variables—when users modify inputs like annual kilometers, the spreadsheet automatically recalculates comparisons, transforming static analyses into personalized decision tools
- A notebook can be built by combining Gemini's synthesis of creator philosophy with actual source material from recent YouTube videos, enabling pattern detection across a creator's content to extract repeatable principles
- The speaker argues NotebookLM should not function as a general chatbot but specifically as a context-bounded tool that eliminates slow initial work and produces polished first versions requiring verification, particularly for calculations, health claims, and financial decisions
Topics
Transcript
[0:00] Notebook LM has long been one of Google's most useful tools, but in recent weeks it has received several features that significantly change how it is used. We're no longer just talking about uploading a PDF, asking questions, and requesting a summary. We already knew that. Now we're talking about taking your sources, your videos, your documents, your data, or even a person's channel and starting to build much more interesting things. For example, you can generate vertical videos from your own sources. You can request files directly from the chat, you can [0:31] use it to analyze real data, and you can turn the content of a creator, a company, or an expert into a kind of private…
Full transcript available for MurmurCast members
Sign Up to AccessMore from Migue Baena IA
Cómo aprender por tu cuenta y volverte imparable con IA
This tutorial demonstrates how to use Gemini Notebook (formerly Notebook LM) to self-teach any skill by curating personal learning sources and converting them into multiple study formats. The four-step process involves selecting quality materials, studying them through different media, teaching the content to others, and integrating learning into daily routines.
Tienes que probar esto antes de crear tus videos con IA 🎬
A new AI video creation tool called 3D Builder within Topw Film Studio allows creators to pre-visualize and plan movie scenes by building interactive 3D environments, positioning characters and objects, controlling camera movements, and reviewing animations before final video generation.
No elijas entre Claude y Codex: úsalos juntos 🤝🏻
The video proposes using Claude and Codex together rather than choosing one over the other to maximize their respective strengths. Claude handles problem analysis and task delegation while Codex executes code generation, allowing users to work around individual API rate limits and leverage complementary capabilities.
Cómo usar Claude mejor que el 99% de la gente
A comprehensive guide to using Claude AI effectively, covering account setup with context import, model selection and reasoning levels, proper prompting techniques, projects and skills for organizing work, and advanced features like Cowork for delegated tasks and Code for building applications.
El archivo de GitHub que deberías de conocer 👀
Anthropic has published a detailed GitHub repository containing a 275,000+ character PRD with nearly 8,000 lines of instructions for one of their models, including specifications on what to do, avoid, and how to use tools. The key value lies not in copying the content but in studying its structure and learning how it organizes rules and defines tool usage for prompt engineering applications.