Google AI Tools | AI Competency Centre
Google AI Tools
Google offers a variety of AI tools which are supported for use at the University of Oxford, including Gemini, NotebookLM, and Google AI Studio. This page covers all of these tools and their features. In summary at Oxford you can access:
- Gemini: Google's main, multimodal AI assistant for generating text, images, and code.
- NotebookLM: A personalised AI research assistant that analyses uploaded documents, notes, and sources.
- Google AI Studio: A prototyping environment for developers to build apps with Gemini.
Google has put together some resources and courses to introduce beginners to their tools. We recommend starting with the 'Getting started with Google AI in Higher Education' course which will help to explain what Google AI is and show practical ways it can support higher education practices.
Accessing Google AI tools
You can use your SSO email (abcd1234@ox.ac.uk) to sign in to the University of Oxford's Google Workspace.
To sign in, follow these instructions:
- Open the page for the tool you wish to use (either gemini.google.com/, notebooklm.google.com/, or aistudio.google.com/)
- Click the Sign in button.
- Enter your Single Sign On (SSO) email address in the format `abcd1234@ox.ac.uk` and press Next. (*Please note that using a department alias such as `@it.ox.ac.uk` will not work*).
- You will be directed to the standard University SSO log-in process.
Note: Signing in with a personal Google account is not protected by enterprise agreements. This includes Pro accounts provided for free by Google as an offer to students. Confidential University data should not be used in personal accounts.
Purchasing Google AI Pro
Following a successful pilot, Google AI Pro will be available to University staff and students through our data-protected workspace at a cost of Ā£170 +VAT per year, with contracts beginning in February. For advanced registration for Google AI Pro, please complete the relevant form below. You will be contacted to confirm your interest before any licences are purchased.
Having trouble with your Google Gemini or NotebookLM?
- Consider first seeking support from the company whose product you're using:
Google Gemini: Google's Gemini Apps Help page offers various pieces of advice and support for using Gemini.
NotebookLM: Googleās NotebookLM Help page covers help topics from getting started to troubleshooting NotebookLM.
If your query is about whether an outage is affecting your connection to a particular tool, try the Google Status page for Google Gemini and NotebookLM - For simple queries regarding University-supported GenAI tools and support, use the AI Competency Centre chatbot to be directed to relevant information.
- For support accessing AI services, contact the Central IT Service Desk.
Google Gemini
Google Gemini provides AI assistance both as a standalone tool and integrated within Google Workspace applications.Ā
What Gemini Does Best
- Educational content creationĀ - Lesson plans, curriculum development
- Grant proposal draftingĀ - Academic and research funding applications
- Document and email summarisationĀ - Processes Google Workspace content
- Research accelerationĀ - Integrates with NotebookLM for knowledge sharing
- Administrative tasksĀ - Streamlines routine educational workflows
Critical Account Risk
High risk of user error:Ā Many users are logged into both personal Gmail and University Google accounts simultaneously. Using Gemini through a personal account providesĀ no protectionĀ andĀ must not be used for University data.Ā
Using Google Gemini
Once you have logged in to Google Gemini, you may wish to learn more about the interface and how to use the tool.
This tab is for users that have no experience or limited experience with Gemini. It covers basic information on the structure of the Gemini interface.
This powerful tool offers revolutionary support for work cases spanning administration, research, education, and beyond. Its chief limitation is user imagination. It is one of the jobs of the AI Competency Centre to help new and experienced users alike explore these possibilities.
Fundamentally, Gemini works by a call-and-response kind of conversational interface. The user sends a request or statement to the AI (called a āpromptā) and then the AI responds to that prompt in a way that resembles a text message conversation.
Gemini Chat Interface
The main chat interface has several major parts. The key elements are labelled below:
- Enter a prompt for Gemini. This is where you click to type in your message and ātalkā to the AI.
- Attach files. You can upload upload directly from your machine or pull them in from Google Drive.
- Tools. Gemini offers a variety of tools that can enhance the conversation with the AI in interesting ways.
- Guided Learning is a feature that facilitates step-by-step learning through guided prompts, hints, and knowledge checks rather than direct answers.
- Deep Research. Deep Research in Gemini functions as an AI-powered research tool. It receives a user-defined prompt and then formulates a multi-step research plan.
- Create an Image. Image creation is a function that lets you ask the AI for a picture and it will create one for you.
- Canvas. Canvas in Gemini is an interactive workspace for document and code creation. It enables real-time collaboration with the Gemini AI. Key functions include generating and refining content drafts. For coding, it generates snippets (e.g., HTML, Python), provides live web app previews, and aids debugging. Content can be exported to Google Docs.
- Model Switcher. This allows you to select the AIās "brain" for the task: Fast is for near-instant everyday tasks, Thinking provides a reasoning layer to double-check logic for tricky problems, and Pro is the most advanced model for heavy-duty coding, math, and complex analysis.
- Microphone. The microphone function in Gemini allows for voice input and interaction. It enables users to speak their queries or commands directly to the AI.
Gemini Extended Interface
Beyond the chat window, Gemini offers additional functionality across the platform.
- Start New chat. This button allows you to start a new conversation from scratch.
- Search chat history. This function allows you to search through all of your past conversations at once, so that you can find all the conversations that relate to a specific topic. This can be super handy if you recall having a conversation about a certain topic, but you can't find it in your chat history.
- My Stuff. This is your personal organized library where Gemini stores all your generated media, research reports, and Canvas projects, allowing you to quickly access and resume previous work.
- Pinned Gems. Gems are ready-to-go conversations with the AI that are already primed with context and background information. You can access prebuilt Gems or create them yourself.
- Chat History. This section catalogues a list of your past conversations with the AI. These will be automatically named based on the content of what you have discussed with the AI.
- Account Information. Here you can click to find information about your account and sign out.
Google Gemini Features
The features available in the University of Oxford Google workspace will be subject to internal governance based on business need and may not be the same as those in consumer versions.
Google Gemini has usage limits in place. This means that the number of prompts, conversations and usage of certain features may be capped.Ā Find out more about usage limits.
| Feature | Standard usage limits | Usage limits with Google AI Pro |
| Model access | Gemini 3 Flash (General access) Gemini 3 Pro & 3 Flash Thinking Up to 100 prompts per day |
Prioritized access to Gemini 3 Pro and 3 Flash Thinking |
| Context window | Up to 32,000 tokens | Up to 1 million tokens |
| Deep Research | Up to 5 reports per month Powered by Gemini 3 Flash |
Up to 20 reports per day Powered by Gemini 3 Flash or Gemini 3 Pro |
| Image generation | Nano Banana: up to 100 images per day Nano Banana Pro: up to 3 images per day |
Nano Banana: up to 1,000 images per day Nano Banana Pro: up to 100 images per day |
| Video generation | Not available | Up to 3 videos per day Using Veo 3 Fast (preview) |
| Workspace integration | Limited | Gemini in Google workspace |
Ā
Models
Within Gemini, users can select between different performance modes depending on task complexity and required depth:
- Fast ā Optimised for quick responses and everyday tasks.
- Thinking ā Designed for more complex problem-solving and structured reasoning.
- Pro ā Allocates more processing time for advanced mathematics, coding, and in-depth analysis
Gems
Gems are specialised versions of Gemini that you can configure for a specific course, role or workflow. You set the instructions, upload relevant files, and choose which tools it can use - creating an assistant that behaves exactly the way you need.
You can get started with Googleās premade Gems like āLearning coachā, āCareer guideā, and āBrainstormerā.
Gemini Canvas
Gemini Canvas is an interactive workspace for drafting, refining, and developing content or code in real time. It supports structured editing, iteration, and export to tools such as Google Docs.
šGemini Canvas
Deep Research
An agentic feature in Gemini that can search across hundreds of websites, synthesise findings, and generate structured, multi-page reports with citations.
Guided Learning
Built in collaboration with educators, cognitive scientists, and learning experts, Guided Learning is an interactive learning mode in Gemini that uses probing, open-ended questions and step-by-step explanations to foster deeper understanding rather than just providing answers.
Video Creation
Create videos in Gemini using Googleās Veo generative video model to transform text or image prompts into short, high-definition cinematic clips. it coverts prompts into 8 second video clips.
Image Creation
Create infographics and images in Gemini using the Gemini 3 Pro Image model to transform text and data into clear, structured visual summaries or eye catching images.
NotebookLM
NotebookLM is an AI-powered research and note-taking assistant grounded exclusively in the sources you upload. It is designed to help academics, researchers, and students organise and analyse large bodies of knowledge. It answers questions and generates outputs, such as summaries, lesson plans, study guides, discussion questions, and quizzes, with in-line citations for accuracy. It supports over 80 languages and can work across multilingual source materials.
What NotebookLM Does Best
- Source-grounded summarisationĀ - Generates summaries directly from uploaded PDFs, text, videos, or websites
- Knowledge synthesisĀ - Builds structured overviews from multiple source documents
- Context-aware Q&AĀ - Answers questions strictly based on your uploaded material
- CollaborationĀ -Ā Allows shared projects with colleagues across the Oxford Google Workspace
- CitationsĀ - Provides inline source references for all generated content
Critical Account Risk
High risk of user error:Ā Many users are logged into both personal Gmail and University Google accounts simultaneously. Using NotebookLM through a personal account providesĀ no protectionĀ andĀ must not be used for University data.Ā
Using NotebookLM
Once you have logged in to NotebookLM, you may wish to learn more about the interface and how to use the tool. Below, we have created various guides to assist with this.
This tab covers basic information on how the tool is structured and how to begin using it.
NotebookLM is an AI-powered research and synthesis environment developed by Google. It enables users to upload documents, extract insights, and generate tailored summaries, outlines, and explanations. It is designed to support work in research, administration, teaching, and creative writing. The core value of NotebookLM is its ability to help users explore and reason through their own materials efficiently.
NotebookLM Homepage Interface
This is the landing page when you first open NotebookLM:
-
Recent Notebooks: Displays notebooks youāve recently opened or worked on.
-
Create New Notebook: A button with a ā+ā symbol to start a new notebook from scratch.
-
Notebook Cards: Each card represents an existing notebook. It shows the title, date last edited, and number of uploaded sources.
-
Settings Icon: Located at the top right, it opens account and app preferences.
-
Account Icon: Displays your Google account. Click to switch or sign out.
Notebook Interface
When a notebook is first created, the workspace is divided into three main panels:
- Sources Panel
- Chat Panel
- Studio Panel
These elements and others are annotated below:
-
Sources Panel (left): ClickĀ AddĀ orĀ DiscoverĀ to upload PDFs, text files, websites, videos, or audio. This is where your reference materials live.
-
Chat Panel (centre): Displays the prompt area and responses. Youāll see āAdd a source to get startedā until sources are uploaded.
-
Studio Panel (right): Offers AI-powered tools such asĀ Audio Overview,Ā Video Overview,Ā Mind Map,Ā Reports,Ā Flashcards, andĀ Quizāall generated based on your sources.
-
Add Note Button (bottom right): Lets you record personal reflections, summaries, or follow-up thoughts from AI responses.
-
Settings and Share: Top-right controls to rename, share, or adjust notebook settings.
For users who are already experienced with generative AI and want more advanced onboarding to Gemini, we have collected the following resources.
NotebookLM features
NotebookLM has usage limits in place. This means that the number of prompts, conversations and usage of certain features may be capped. Find out more about usage limits.
| Feature | Standard usage limits | Usage limits with Google AI Pro |
| Notebooks | Up to 100 notebooks | Up to 500 notebooks |
| Sources | Up to 50 sources per notebook | Up to 300 sources per notebook |
| Queries | Up to 50 daily chat queries | Up to 500 daily chat queries |
| Deep Research | Deep Research: 10 per month | Deep Research: 20 per day |
| Audio overviews | Audio overviews: 3 per day | Audio overviews: 20 per day |
| Video overviews | Video overviews: 3 per day | Video overviews: 20 per day |
NotebookLM Studio
Studio is NotebookLMās creation space for turning your uploaded materials into a range of outputs. From a single notebook, you can generate audio and video overviews, mind maps, structured reports, and learning materials such as quizzes and flashcards ā all based only on the sources you provide.
Mind Maps
Mind Maps create an interactive visual overview of the key concepts, themes, and entities across all sources in your notebook. They surface relationships between ideas and allow you to explore how topics connect across multiple documents.
Selecting any node reveals how that concept is discussed within your uploaded materials, helping users trace themes, compare perspectives, and identify patterns across sources.
šUse Mind Maps
Audio Overviews
Audio Overviews convert your uploaded materials into a conversational, podcast-style summary voiced by two AI hosts. NotebookLM synthesises notes, readings, research, webpages, or videos into an accessible audio format.
Using the customise option, you can direct the focus of the discussion to prioritise specific themes, questions, or sections of your sources.
Video Overviews
Video Overviews transform your uploaded sources into a short, illustrated video summary presented by AI hosts. They provide a visual way to review key ideas and quickly grasp the main points of longer materials.
NotebookLM can generate a video overview from lesson plans, research papers, articles, or other uploaded content, helping to communicate complex material in a concise, accessible format.
Reports
Reports transform your uploaded sources into structured, functional documents. You can generate formats such as briefing documents, study guides, literature reviews, blog posts, research proposals, timelines, or fully custom outputs.
NotebookLM synthesises key themes, evidence, and insights from your materials into a coherent draft, helping you move from reading and research to usable outputs more efficiently.
Flashcards and Quizzes
Flashcards and quizzes generate revision materials directly from your uploaded sources. They extract key terms, concepts, and questions to support retrieval practice and knowledge checks.
Infographics
Infographics transform your uploaded documents into clear, structured visual summaries. They synthesise key points, data, and relationships into maps, timelines, flowcharts, or comparative layouts.
All visuals are generated solely from your notebook sources, supporting accuracy while helping to communicate complex material in a concise, accessible format.
Slide decks
Slide decks turn your sources into a presentation which can be presented instantly within NotebookLM or shared as a PDF.
Google AI Studio
Google AI Studio is a web-based development environment for working directly with Googleās Gemini models.
It provides a structured space to design, test and refine prompts, explore model behaviour, and prepare AI features for integration into applications.
AI Studio includes aĀ PlaygroundĀ for testing prompts and model settings, and aĀ BuildĀ environment for preparing API integrations and exporting code.
What you can do
With Google AI Studio, you can:
- Access Gemini models for text, code, reasoning and multimodal tasks
- Experiment with prompt design and structured outputs
- Adjust model settings such as temperature and response length
- Generate and manage API keys for development use
- Export code for integration into applications
Who it is for
Google AI Studio is primarily intended for developers, technical teams and researchers who require direct control over model configuration and testing.