Starts Nov 23Live Cohort 3 of Vibe Coding & AI Agents: 3 weeks, 9 live hands-on sessionsCohort 3 (Nov 23) →
ReleaseInsight · On RadarKDnuggets→ original

Google expands NotebookLM with five new features for advanced research work

Google has added five features to NotebookLM at once, aimed at people who work with long-form research and presentations. Users can now edit individual slides with a prompt, export decks to PPTX, build artifacts directly from chat, generate video overviews, and upload EPUB books without splitting them up by hand. The update moves NotebookLM away from being a note-taking app and closer to a full environment for analyzing and packaging results.

AI-processed from KDnuggets; edited by Hamidun News
Google expands NotebookLM with five new features for advanced research work
Source: KDnuggets. Collage: Hamidun News.
Listen to article
Article Fact Sheet & Key Figures
ParameterDetails
Primary sourceKDnuggets ↗
Editor-in-Chief
Key metric #1Upload EPUB, PDF, CSV, and internal documents into a single notebook
Key metric #2Use chat to quickly test hypotheses and build a narrative
Key metric #3Turn conclusions into a brief or slide deck directly from the conversation

Google has expanded NotebookLM, and it is clearly targeting users who treat the service as a working environment for research, presentations, and internal documents rather than as a study aid. The latest batch of updates adds targeted slide editing, PPTX export, artifact generation from chat, video overviews, and EPUB support.

More precise slide work

One of the most practical additions is targeted editing of individual slides via a text instruction. Previously, a mistake in a single chart or line of text meant regenerating the entire deck. Now you can open a finished presentation in Studio and ask NotebookLM to change only the part you need: fix a metric, turn a list into a comparison table, or put the emphasis on a specific conclusion. For teams that polish presentations over several iterations, this eliminates unnecessary regeneration.

"Update the 2025 revenue using the value from Table 2 and add the source in a footnote."

The second important detail is exporting slides to PPTX. NotebookLM remains the place where you build the draft and the logic of a presentation, while PowerPoint once again serves as the final container for corporate work. According to the feature description, the structure and visual layout carry over into the standard format without manually copying each slide. These decks may still need some cosmetic touch-ups, but the path from research to presentation is now noticeably shorter, especially for analysts, product managers, and consulting teams.

From chat to output

Another update removes an extra switch between modes: you can now create artifacts directly inside the chat. If a conversation with NotebookLM has already shaped the logic of a document or explanation, you just ask the system to turn that thread into a slide deck, a two-page brief, or another output. This helps when the strongest framing emerges not from a pre-planned template but from a live dialogue about data, hypotheses, and follow-up questions.

Google is also beefing up video overviews. NotebookLM can now produce more cinematic videos from a notebook's materials, combining script, structure, and visual presentation in a single pipeline. For internal demos, briefings for leadership, or onboarding, this is a notable step forward: you no longer need to write a script separately, assemble a storyboard, and manually condense a long piece of research into a short explainer. Judging by the description, the key requirement is carefully prepared sources and a clear prompt tailored to a specific audience.

Long sources without the pain

EPUB support makes NotebookLM noticeably more useful for people who work not only with notes and PDFs but also with full books, technical manuals, and long playbook documents. Instead of manually splitting and reconverting files, you can upload a long source as is and then cross-reference it with spreadsheets, internal documents, and other files in a single notebook. This fits especially well with research and data-driven workflows that require linking book chapters to figures, policies, and operational metrics.

The updated NotebookLM workflow now looks like this:

  • Upload EPUB, PDF, CSV, and internal documents into a single notebook
  • Use chat to quickly test hypotheses and build a narrative
  • Turn conclusions into a brief or slide deck directly from the conversation
  • Make targeted edits to individual slides without rebuilding the whole deck
  • Export the result to PPTX or produce a video overview for stakeholders

Taken together, the service is moving further away from the role of a "smart summary" tool and becoming an environment where you can cover almost the entire path from a raw pile of sources to a finished artifact. For power users, this essentially means fewer context switches: less manual packaging, less copy-pasting, less regeneration because of one wrong number or a bad slide, and a more predictable route from analysis to final output.

What this means

NotebookLM is starting to compete not only with AI note-taking assistants but also with pieces of the familiar work stack, from research tools to presentation software. If Google gets export quality and video formats right, the service could become a genuine intermediate operating system for knowledge within teams.

Story

Key Questions Answered

What are the key details on More precise slide work?

One of the most practical additions is targeted editing of individual slides via a text instruction. Previously, a mistake in a single chart or line of text meant regenerating the entire deck. Now you can open a finished presentation in Studio and ask NotebookLM to change only the part you need: fix a metric, turn a list into a comparison table, or put the emphasis on a specific conclusion. For teams that polish presentations over several iterations, this eliminates unnecessary regeneration.

What are the key details on From chat to output?

Another update removes an extra switch between modes: you can now create artifacts directly inside the chat. If a conversation with NotebookLM has already shaped the logic of a document or explanation, you just ask the system to turn that thread into a slide deck, a two-page brief, or another output. This helps when the strongest framing emerges not from a pre-planned template but from a live dialogue about data, hypotheses, and follow-up questions. Google is also beefing up video overviews.

What are the key details on Long sources without the pain?

EPUB support makes NotebookLM noticeably more useful for people who work not only with notes and PDFs but also with full books, technical manuals, and long playbook documents. Instead of manually splitting and reconverting files, you can upload a long source as is and then cross-reference it with spreadsheets, internal documents, and other files in a single notebook. This fits especially well with research and data-driven workflows that require linking book chapters to figures, policies, and operational metrics.

ZK
Zhemal Khamidun — CPO at Alpina Digital, product manager of AlpinaGPT. Building AI products since 2019. Hamidun News is a pet project that grew into a production aggregator with 6,000+ readers.
✦ Unified AI Workspace · Nomad

Test Frontier LLMs & Autonomous Agents in One Workspace

Run Claude Opus 5.5, GPT-6, Gemini 3.1, and DeepSeek side-by-side in Nomad AI (nomadnet.ai) — unified model routing, deep research, and multimodal workflows under one subscription.

Try Nomad AI →Launch Nomad Agent
What do you think?
Loading comments…