This evergreen history article uses authoritative archives and official records. Exact dates are used when documented; gradual inventions and rollouts are described as periods rather than being assigned a misleading single birthday.
Quick facts
- Google introduced the experiment as Project Tailwind at Google I/O in May 2023.
- It was renamed NotebookLM in July 2023 and built around user-provided sources rather than open-ended chat alone.
- NotebookLM can summarize, compare and explain information from documents, websites, videos and other supported sources.
- Audio Overviews turned notebook material into an AI-generated conversational discussion and became a signature feature.
- In July 2026 Google announced that NotebookLM would become Gemini Notebook as notebook capabilities moved deeper into Gemini.
Origins and founding
Project Tailwind began inside Google Labs as an experiment in personalized, source-grounded AI. The idea was simple but important: a user would provide a collection of material, and the model would act as a research assistant specifically for that collection. This reduced the need to repeat context in every prompt and made citations to the source material central to the experience.
Google showed Tailwind at I/O 2023, initially describing it as an AI-first notebook for students. The project was soon renamed NotebookLM, with “LM” referring to language model. The broader name reflected use by researchers, writers, teachers, journalists, businesses and anyone trying to understand a large set of documents.
The product takes shape
NotebookLM opened gradually, first as an experimental US product and later in more countries. Users could upload Google Docs, PDFs and copied text, then ask questions or generate summaries and study materials. Answers included references that helped users check the underlying source instead of accepting unsupported prose.
The product’s most visible breakthrough was Audio Overviews. It transformed a notebook into a lively discussion between AI-generated hosts. Although the format could sound like a podcast, its real value was compression: long source collections could become an accessible spoken explanation. Google later added controls, multiple languages and richer overview formats.
Technology and major features
NotebookLM uses Gemini models together with retrieval over the user’s selected sources. The system divides material into searchable representations, finds passages relevant to a question and provides those passages as context for the model. This approach is commonly called retrieval-augmented generation. It does not guarantee perfection, but it helps keep answers connected to the notebook.
Features expanded to include mind maps, study guides, briefing documents, FAQs, timelines, slide-style and video summaries, and collaboration. Supported sources grew beyond documents to include websites, YouTube transcripts and other media. Notebook organization became increasingly important as people built persistent research spaces rather than one-time chats.
Growth and wider influence
NotebookLM popularized a distinct category of AI product: the grounded personal research workspace. Students used it to revise courses, companies used it to understand policy libraries, and creators used it to analyze interviews or source material. Audio Overviews also influenced competing products and changed how people expected AI to present complex information.
The service strengthened Google’s larger Gemini strategy. Instead of treating every AI interaction as a blank chat, notebook spaces preserved a controlled body of evidence. This model is useful whenever accuracy, traceability and repeated work with the same material matter.
Challenges, criticism and responsibility
A source-grounded system can still misunderstand a passage, omit context or confidently combine facts incorrectly. Users must check citations, especially for medical, legal, financial or academic decisions. Copyright and privacy also matter because uploaded files may contain protected or confidential information.
Audio and video summaries can make uncertain interpretation sound polished and authoritative. The conversational style may hide simplification. Organizations therefore need clear data-handling rules, and students should use the tool to support learning rather than submit generated analysis without verification or attribution.
Where it stands in 2026
By July 2026, Google was moving NotebookLM’s identity and capabilities into the wider Gemini experience under the name Gemini Notebook. The transition reflected a strategy in which notebooks, source grounding and multimodal research became native parts of Google’s main AI assistant rather than a separate experiment.
The core idea remains influential: users should be able to create a durable AI workspace around their own trusted material. Future versions are likely to support larger libraries, stronger collaboration, richer media production and more transparent evidence tracking across long research projects.
Timeline
| Year | Location | Event | Why it mattered |
|---|---|---|---|
| May 2023 | Mountain View, United States | Google reveals Project Tailwind at I/O | Introduced an AI notebook grounded in user-selected sources. |
| July 2023 | United States | Project Tailwind becomes NotebookLM | Established the product’s public identity as a language-model notebook. |
| 2024 | International markets | Audio Overviews and broader availability arrive | Made source collections easier to understand through conversational audio. |
| 2025 | Global | More source types, languages and study formats launch | Expanded NotebookLM into a multimodal research workspace. |
| July 2026 | Global Gemini ecosystem | Google announces the Gemini Notebook transition | Moves notebook capabilities deeper into the main Gemini product family. |
Frequently asked questions
What was NotebookLM originally called?
It was first announced as Project Tailwind in May 2023.
What makes NotebookLM different from a normal chatbot?
It is designed to answer from a collection of sources selected by the user and to provide references back to those sources.
What are Audio Overviews?
They are AI-generated spoken discussions that explain and connect information from a notebook’s sources.
Is NotebookLM becoming Gemini Notebook?
Google announced in July 2026 that NotebookLM would transition to the Gemini Notebook name as the capability became more integrated with Gemini.
Final perspective
NotebookLM proved that a useful AI assistant does not always need access to everything; sometimes it becomes more valuable when deliberately limited to a trusted set of sources. Its transition into Gemini Notebook shows how source-grounded workspaces are becoming a standard layer of modern AI productivity.