How-To

Google NotebookLM in the UK: how to use Google’s AI research tool in 2026

Google NotebookLM is a free, grounded AI research tool that answers only from sources you upload. Our UK guide covers features, pricing in GBP and how to start.

Google NotebookLM is the search giant’s grounded AI research assistant, a tool that answers questions using only the documents, links and notes you feed it, and in 2026 it has quietly become one of the most useful free apps a UK student, researcher or small-business owner can have. Unlike a general chatbot that draws on the whole internet, NotebookLM works strictly from your own uploaded sources, so its answers come with citations you can click straight back to the page or paragraph they came from. This guide explains, in plain British English, exactly what the tool does, how its Audio Overviews and Video Overviews work, what the free and paid tiers cost in pounds, what Google says about your data, and a step-by-step way to get going today.

Key facts
  • NotebookLM is a source-grounded research assistant powered by Google’s Gemini models: it answers only from the sources you upload and cites them (notebooklm.google).
  • The free tier allows up to 100 notebooks, 50 sources per notebook, 50 chats a day and 3 Audio Overviews a day (support.google.com).
  • NotebookLM Plus, available through a Google AI Plus, Pro or Ultra subscription or eligible Google Workspace plans, raises those to 200 notebooks, 100 sources, 200 chats and 6 Audio Overviews a day.
  • UK pricing: Google AI Plus is £4.49 a month, Google AI Pro is £18.99 a month and Google AI Ultra starts at £79.99 a month (gemini.google, June 2026).
  • Google states that, for personal accounts, it does not use your uploaded NotebookLM content to train its models unless you choose to give feedback; Workspace data is never used for training (support.google.com).

What Google NotebookLM actually is

At its heart, the tool is a research notebook with an AI assistant bolted on, and the crucial difference from ChatGPT or a standard Gemini chat is that it is grounded. You give it sources, and it confines itself to them. According to Google’s own product pages on notebooklm.google, you can upload PDFs, Google Docs and Slides, copied text, website links and even YouTube videos, and the assistant then reads everything and becomes an expert on that specific material. Ask it a question and it answers from your documents, footnoting each claim with an inline citation that jumps you to the exact source passage. For anyone who has ever worried about an AI inventing facts, that grounding is the whole point: if it is not in your sources, NotebookLM is designed not to say it.

That makes it a very different proposition from the general assistants we have covered before. If you are weighing up the broader AI landscape, our verdict on whether Gemini is worth it in the UK and our comparison of Copilot versus Gemini are useful context, but NotebookLM sits in its own category. It is not trying to chat about anything under the sun; it is trying to help you understand a defined set of material faster. Google first launched it as an experiment and has since expanded it into a polished product, available free at notebooklm.google.com to anyone with a Google account.

Students using Google AI research tools to learn in the AI era
Image: Google

Audio Overviews and Video Overviews explained

The feature that turned NotebookLM into a viral hit is the Audio Overview. With a couple of clicks it generates a podcast-style discussion between two AI hosts who talk through the key points of your sources, complete with the natural back-and-forth of a real chat show. It is genuinely useful for revision or for getting across a dense report on a commute, and Google has since added the ability to interact with the hosts using your voice, asking them to slow down, explain a concept differently or dig into a particular point. Google has also expanded Audio Overviews to more than 80 languages, moving from short clips to full-length deep dives, as detailed on the official blog.google.

The newer companion feature is the Video Overview. Rather than audio alone, it produces narrated slides, with the AI host building visuals, pulling in images, diagrams, quotes and figures from your documents to illustrate the points it is making. Google describes a more cinematic version with fluid animations for visualising complex narratives and academic research. Both Audio and Video Overviews now live alongside Mind Maps and Reports in a redesigned Studio panel, and you can keep several outputs of each type in a single notebook. The official Google video below introduces Video Overviews and shows the format in action.

NotebookLM updates shown at Google I/O 2026
Image: Google

Free versus paid: source limits and UK pricing

The good news for most readers is that NotebookLM is genuinely useful for free. On the standard free tier, Google’s support pages list a generous allowance: up to 100 notebooks per user, 50 sources per notebook, 50 chat messages a day and 3 Audio Overviews a day. For a student writing one essay or a sole trader summarising a handful of contracts, that is usually plenty. The paid upgrade, NotebookLM Plus, roughly doubles the headroom, raising the limits to 200 notebooks, 100 sources per notebook, 200 chats a day and 6 Audio Overviews a day, along with more capacity for the newer Studio formats such as data tables and slide decks.

NotebookLM Plus is not sold on its own. You get it as part of a wider Google AI subscription, or through eligible Google Workspace plans for businesses and schools. In the UK, as listed on gemini.google in June 2026, Google AI Plus costs £4.49 a month, Google AI Pro is £18.99 a month, and Google AI Ultra starts at £79.99 a month with a higher-limit tier at £189.99 a month. All of these bundle the higher NotebookLM allowances alongside extras such as more access to Google’s latest Gemini models. If you are comparing the assistants more broadly, our breakdowns of Gemini UK pricing, ChatGPT UK pricing and Copilot UK pricing set the wider market in pounds so you can see where each tier sits.

LimitFree (standard)NotebookLM Plus
Notebooks per user100200
Sources per notebook50100
Chats per day50200
Audio Overviews per day36
Source: NotebookLM Help (support.google.com), June 2026.
Google AI education partnership supporting study and revision in schools
Image: Google
Video: Google

The video above from Google’s official channel introduces Video Overviews, the narrated-slides format that turns your sources into a short visual explainer. It is a good way to see how the tool reuses the diagrams and quotes inside your own documents rather than generating generic stock content. That fidelity to your material is what separates NotebookLM from a general AI video generator, and it is also why the tool has caught on so quickly with teachers, analysts and anyone who has to digest long documents under time pressure.

What Google says about your data and privacy

Privacy is the question we get asked most, and here Google is reasonably clear. According to its NotebookLM Help pages, the sources you upload stay private unless you deliberately choose to share a notebook with someone. For personal Google accounts, Google states that it does not use your uploaded content to train its AI models, with one stated exception: if you actively submit feedback, such as a thumbs-up or thumbs-down on a response, that flagged content can be reviewed by trained teams to fix problems. For Google Workspace accounts, the protection is stronger still: uploads, queries and model responses are not reviewed by humans and are not used to train models, even when you provide feedback.

For UK readers that is broadly reassuring, but a couple of caveats are worth stating plainly. “Not used for training” is not the same as a guarantee of permanent deletion, and if you do hit the feedback buttons on a personal account, you are opting that specific exchange into human review. The sensible rule for anything genuinely confidential, whether that is client data, medical notes or commercially sensitive figures, is to use a Workspace account where available and to avoid submitting feedback on those notebooks. Privacy settings are a recurring theme across Google’s AI products, and our guide to the Gemini app privacy settings UK users should check walks through the wider controls in the same spirit.

Google AI research direction shaping the future of AI tools like NotebookLM
Image: Google

Step by step: how to create your first notebook

Getting started takes minutes. First, head to notebooklm.google.com and sign in with your Google account. Click “Create new” to open a fresh, empty notebook, then give it a clear name so you can find it later, such as “GCSE biology revision” or “Q3 supplier contracts”. The interface is split into three panels: your sources on the left, the chat in the middle, and the Studio panel on the right where Audio Overviews, Video Overviews, Mind Maps and Reports live. There is nothing to install; it runs entirely in your browser, and there is a companion mobile app for listening on the go.

Next, add your sources. Click “Add source” and choose how you want to bring material in: upload a PDF, pick a Google Doc or Slides file from your Drive, paste a website URL, drop in a YouTube link so the tool reads the transcript, or paste raw text. You can add up to 50 sources per notebook on the free tier, so a typical project might combine a few PDFs, a couple of web articles and a lecture recording. Once the sources are in, NotebookLM reads and indexes them, and a short summary of each appears so you can confirm it has understood the material before you start asking questions.

Asking questions and generating an Audio Overview

With sources loaded, the chat box is where the real work happens. Type a question in plain English, such as “What are the three main arguments against the proposal?” or “Summarise the payment terms across all the contracts”, and the assistant answers using only your documents. Crucially, every answer carries numbered citations: click one and NotebookLM jumps you to the exact passage in the exact source, so you can verify the claim yourself in seconds. That citation trail is what makes the tool trustworthy for serious work, because you are never asked to take its word on faith. You can also ask follow-up questions, request a study guide, a briefing document, a timeline or a table of key terms, and save any of these into the notebook.

To create an Audio Overview, open the Studio panel and click the Audio Overview tile. Within a few minutes the tool produces a podcast-style conversation about your sources that you can play, download or, on the latest version, interrupt with your voice to ask the hosts for clarification. For revision this is transformative: you can turn a dense textbook chapter into a 10-minute discussion to listen to while cooking. To generate a Video Overview instead, choose that tile and NotebookLM builds narrated slides using the visuals inside your documents. If you work across Google’s productivity apps too, our walkthrough on how to use Gemini in Gmail and Docs pairs neatly with this, since the two tools cover different jobs.

Study, revision and small-business use cases

For students, NotebookLM is close to a purpose-built revision machine. Load your lecture notes, the set texts and a few past papers, then ask it to generate a study guide, quiz you on key concepts, or build a mind map that shows how topics connect. The Audio Overview turns the same material into something you can revise hands-free, and because every answer is cited, you can always trace a point back to the original source rather than memorising something the AI invented. Google has leaned hard into education, with school and university partnerships and a clear pitch to learning in the AI era, and the tool’s grounding is exactly what makes it safer for study than an open-ended chatbot.

For UK small businesses, the use cases are just as concrete. A consultant can drop a stack of client reports into a notebook and ask for a one-page brief. A shop owner can upload supplier contracts and ask which renews first or where the penalty clauses sit. A marketing team can feed in past campaign decks and generate a fresh summary deck in the Studio panel. Because it works only from your own material, it is well suited to the kind of focused, document-heavy tasks that general assistants handle clumsily. The same appetite for practical AI is driving tools across the sector, as our coverage of Meta Business AI for UK small businesses and Microsoft 365 Copilot for UK small business shows. For the bigger strategic picture, the views of leaders like Anthropic’s Dario Amodei on the UK are worth following, and on the public-sector side our look at the HMRC AI tax system deal with Microsoft shows how grounded AI is reaching government too.

Google AI updates from May 2026 including research and study tools
Image: Google

Tips to get more out of the tool

A few habits make a big difference. Keep each notebook tightly focused on one project rather than dumping everything into one place, because the assistant gives sharper answers when the sources are coherent. Name your sources clearly, since the citations read better when “Annual report 2025” beats “document (3).pdf”. When you ask questions, be specific about the format you want, asking for a bulleted summary, a table or a timeline, and the tool will oblige. And do skim the auto-generated source summaries before you rely on a notebook, just to confirm nothing has been mis-read or truncated on upload.

The single feature that makes NotebookLM trustworthy for real work is the citation: every answer points back to the exact passage in your own sources, so you are never asked to take the AI on faith.

Mobile Tech World

Finally, treat the Studio outputs as drafts, not gospel. The Audio and Video Overviews are excellent for getting across material quickly, but for anything you will publish or present, read the underlying sources too. The strength of NotebookLM is that it always lets you do exactly that, with the citations one click away.

Where to start or check pricing in the UK

The simplest first step costs nothing: visit notebooklm.google.com, sign in, and build a notebook around a project you already have on the go. The free tier is more than enough to decide whether the tool earns a place in your workflow. If you find yourself bumping into the limits, the upgrade path runs through Google’s AI subscriptions, and you can check current pricing and what each tier unlocks at gemini.google. For UK buyers, weigh the £4.49-a-month Google AI Plus against the £18.99 Pro and £79.99 Ultra tiers, and remember that those plans bundle wider Gemini access on top of the higher NotebookLM allowances, so the value depends on how much of the rest you will use.

If you are deciding between assistants rather than committing to one ecosystem, our comparison of Claude versus Copilot versus Gemini for UK business and the wider Gemini features UK users actually get from Google I/O 2026 will help you see where NotebookLM fits in Google’s stack. For most people the answer is simple: it does not replace a general chatbot, it complements one, and as a free, grounded research aid it is hard to beat.

Our verdict

NotebookLM is one of the rare AI tools that is both genuinely free and genuinely useful, and for UK students and small-business owners it is close to essential. The grounding is the whole story: by answering only from the sources you give it and citing every claim, it sidesteps the trust problem that dogs general chatbots, and the Audio and Video Overviews turn dense material into something you can actually absorb. The free tier, with 100 notebooks and 50 sources each, will cover most people comfortably, and the paid Plus tier through Google AI Plus, Pro or Ultra is there for heavier users who need more headroom. On privacy, Google’s stated position that it does not train on your personal-account uploads unless you give feedback is reassuring, though anyone handling confidential data should lean on a Workspace account and skip the feedback buttons. Our advice is straightforward: spend ten minutes building a notebook around a real project this week. If it sticks, it will quietly become one of the tools you reach for most, and you will not have spent a penny to find out.

What is Google NotebookLM?

Google NotebookLM is a source-grounded AI research assistant powered by Google’s Gemini models. You upload your own documents, links, slides or YouTube videos, and it answers questions using only that material, citing the exact passage each answer comes from. It is free to use at notebooklm.google.com with a Google account, with a paid Plus tier for heavier use.

Is NotebookLM free in the UK?

Yes. The standard free tier gives you up to 100 notebooks, 50 sources per notebook, 50 chats a day and 3 Audio Overviews a day, which is plenty for most students and small businesses. NotebookLM Plus, with double those limits, comes bundled with a Google AI Plus (£4.49 a month), Pro (£18.99) or Ultra (from £79.99) subscription or eligible Google Workspace plans.

What are Audio Overviews and Video Overviews?

Audio Overviews are podcast-style discussions between two AI hosts that summarise your sources, and you can interrupt them with your voice to ask for clarification. Video Overviews are narrated slides that build visuals from the images, diagrams and quotes inside your documents. Both are generated in the Studio panel alongside Mind Maps and Reports.

How many sources can I upload to NotebookLM?

On the free tier you can add up to 50 sources per notebook, across PDFs, Google Docs and Slides, website links, YouTube videos and pasted text. NotebookLM Plus raises this to 100 sources per notebook. You can keep up to 100 notebooks on the free tier and 200 on Plus.

Does Google use my NotebookLM data to train its AI?

For personal Google accounts, Google states it does not use your uploaded content to train its models unless you actively submit feedback, such as a thumbs-up or thumbs-down, which can then be reviewed by trained teams. For Google Workspace accounts, your uploads and queries are not reviewed by humans and are not used for training even with feedback.

How do I create a notebook in NotebookLM?

Go to notebooklm.google.com, sign in with your Google account, and click “Create new” to open an empty notebook. Name it, then click “Add source” to upload PDFs, pick Google Docs from Drive, paste website or YouTube links, or paste text. Once the sources are indexed, type questions in the chat and the assistant answers with citations.

Is NotebookLM good for revision and study?

Yes, it is one of its strongest uses. Load lecture notes, set texts and past papers, then ask it to build a study guide, quiz you, or generate a mind map. The Audio Overview turns the material into a hands-free discussion for revising on the go, and every answer is cited so you can trace points back to the original source rather than trusting an AI’s memory.

Can small businesses use NotebookLM?

Absolutely. A consultant can summarise client reports into a one-page brief, a shop owner can query supplier contracts for renewal dates or penalty clauses, and a marketing team can turn past decks into a fresh summary. Because it works only from your own material and cites everything, it suits focused, document-heavy tasks that general chatbots handle clumsily.

Related reading on MTW

Buyer action

Where to buy or check next

Use this as the final check before ordering a phone, changing network or trusting a headline monthly price.

Stay in the loop

Get MTW reporting, reviews, guides, and buying advice in your inbox.

Subscribe

Reader discussion

Leave a comment

Comments are moderated. Keep it useful, accurate, and on topic.

Join the discussion

Your email address will not be published. All comments are held for moderation.

Spam protection

Keep reading

Today on MTW

The latest stories moving through the newsroom.

Keep reading

Latest reviews

Recent hands-on verdicts and product reads.

Keep reading

Buying guides

Practical UK buying advice and comparisons.

Keep reading

From the archive

Legacy reporting from the MobileTechWorld back catalogue.