Unless, that is, you have access to Canva Pro. There is a workaround that gets you from a locked NotebookLM PDF to a fully editable set of slides, and it is more straightforward than you might expect.

Here is how it works.

What You Will Need

You need a NotebookLM slide deck that you have already created, and a Canva Pro account. If your school uses Canva for Education, check whether your account includes access to premium features (those marked with a small crown icon). Most full Canva for Education accounts do include these, but it is worth confirming before you start.

Step One: Download Your NotebookLM Slides as a PDF

Open your slide deck in NotebookLM. Click the three dots in the top right corner of the presentation and select the option to download as a PDF. The file will save to your device in the usual way. This is your starting point.

Step Two: Import the PDF into Canva

Switch over to Canva. You do not need to create a new project first. Simply drag your PDF file directly onto the Canva home screen or into an open canvas and Canva will upload and open it automatically. Each slide from your presentation becomes a separate page within Canva.

Step Three: Use the Grab Text Feature

This is where the Canva Pro account earns its place. Navigate to the page you want to edit and click on it. Along the left-hand panel, select the option to see all tools. Scroll down until you find “Grab Text.” This is a premium feature, so look for the crown icon to confirm it is available to you.

Once you click Grab Text, you have two options. You can click on individual blocks of text to make just those sections editable, or you can select “All Text” to convert every piece of text on that page at once. Click Grab and Canva will separate out the text elements from the background graphic, making each one independently editable.

From there, you can click into any text block and change it as you would in any design tool. Correct a label, update a date, swap out a term that does not quite match the vocabulary your class has been using. Any changes you make are saved within your Canva project.

A Note on the Process

This works on a page-by-page basis, so if you have a long deck you will need to apply the Grab Text step to each slide individually. It adds a little time, but for a deck you are planning to use more than once, or one where accuracy really matters, it is well worth the extra few minutes.

It is also worth noting that the underlying graphic on each slide stays intact. You are editing the text layer on top of it, not redesigning the slide itself. The visual style of the NotebookLM output is preserved, which is actually one of the advantages of this approach.

Classroom Application

1. Adapt slides for your specific class. NotebookLM generates content from your sources, but the vocabulary or level of detail might not be quite right for your year group. Once your slides are editable in Canva, you can adjust the language to match what your pupils have already encountered, without starting from scratch.

2. Correct subject-specific terminology. AI-generated content occasionally uses terminology that is slightly off for the UK curriculum, or that does not match the phrasing in your scheme of work. This workflow gives you a quick way to fix those inconsistencies before the slides go in front of a class.

3. Create differentiated versions. With editable text, you can duplicate your Canva project and produce a simplified version of the same slide deck for pupils who need more accessible language, or an extended version for those who need more detail. One source deck, multiple outputs.

Notebook LM slides finally editable with Canva
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

This isn’t about squeezing every last drop of productivity out of an AI. It’s about knowing what’s available so you can make an informed choice about whether it’s worth your time.

Here are six features that are worth knowing about.

1. Projects: Set It Up Once, Reuse It Every Time

One of the more frustrating things about using AI tools is that every new conversation starts from zero. You find yourself typing the same contextual information each time: your year group, your curriculum framework, your school’s policies. It gets old quickly.

The Projects feature addresses this. You can create a Project as a kind of permanent workspace, uploading the documents that underpin your planning: schemes of work, curriculum overviews, lesson plan templates. You can also add custom instructions, things like “always suggest differentiation for SEND pupils” or “use the school’s agreed marking language.”

Once that’s set up, every conversation you have inside that Project already has that context baked in. You don’t have to re-explain yourself.

Try it: Start a lesson planning conversation with “Plan a lesson on [topic] using my uploaded scheme of work.” The difference in the output is usually noticeable straight away.

2. Artefacts: From Text to Usable Tool

Sometimes Claude produces a perfectly decent response, and you still feel slightly stuck because it’s a wall of text when what you actually needed was a table, a formatted template, or an interactive quiz. That gap between “the AI wrote something” and “I can use this in my classroom” is where a lot of time gets lost.

Artefacts is designed to help with this. When it’s enabled, Claude can produce visual, formatted outputs in a side panel: a coded quiz you can actually interact with, a structured email template, a tracker laid out as a table rather than a list.

Try it: Describe the lesson you’ve just taught and ask Claude to produce a three-question exit ticket as an Artefact. It’s a small thing, but it saves the reformatting step.

3. Custom Style: Writing That Sounds Like You

If you’ve ever read a piece of AI-generated writing and thought “this sounds like nobody in particular,” that’s a fair criticism. The default output can be generic in tone, which is fine for internal planning but less ideal for newsletters, reports, or parent communications where your voice matters.

Custom Style allows you to upload samples of your own writing so Claude can learn how you write. Are you fairly direct? Do you tend towards warmth and encouragement? Do you use specific vocabulary or sign-off phrases? Once you’ve trained it on your examples, the outputs will reflect your style rather than a neutral default.

Try it: Paste in a few paragraphs from a previous newsletter or report, then ask Claude to draft something new in that style. Compare it to a standard output and see which one actually sounds like you.

4. Extended Thinking: Slower, More Considered Responses

For quick tasks, a fast response is exactly what you want. For more complex decisions, it can actually be unhelpful to get an instant answer without any visible reasoning behind it.

Extended Thinking is a mode that asks Claude to work through its reasoning before giving you a response, effectively showing its workings. This is particularly useful when the stakes are higher: analysing assessment data, thinking through an intervention strategy, or working out how to handle a difficult professional situation.

Try it: Upload some anonymised pupil data and ask Claude, using Extended Thinking, to identify which students may need additional support in a specific area and to suggest some evidence-informed strategies. The depth of response is usually significantly better than a standard prompt.

5. File Upload: Analyse Documents Without the Copy-Paste Nightmare

Most teachers have at some point tried to copy large amounts of text from a PDF into an AI chat window, reformatted it several times, and then given up. The File Upload feature bypasses all of that.

Claude can read PDFs, Word documents, spreadsheets, and more. This means you can hand it a lengthy curriculum document, an Ofsted framework, or a class data spreadsheet and ask specific questions about it rather than trying to summarise it yourself first.

Try it: Upload a dense policy or framework document and ask Claude to summarise the changes that will affect your year group or subject next term. It’s a reasonable use of ten minutes.

6. Forking Conversations: Try a Different Angle Without Starting Over

There’s a particular frustration familiar to most people who use AI tools regularly: you get a response that’s almost right, but not quite. Maybe the tone is slightly off, or it’s focused on the wrong aspect. Your options have usually been to write a follow-up prompt and hope for the best, or delete everything and start from scratch.

The conversation forking feature changes this. You can edit any previous message in a conversation, and rather than overwriting what came before, it branches from that point. You can compare both versions and keep whichever works better.

Try it: Ask Claude to explain a concept for your class. If the response isn’t quite right, edit your original prompt to emphasise a different angle, such as using analogies, or a different reading level. Compare the two outputs side by side.

Classroom Application

If you want to start putting some of this into practice, here are three straightforward steps:

1. Set up one Project this week. Pick one unit of work you’re currently teaching and upload the relevant planning documents. Add a short custom instruction about your class or your school’s preferred approach. Use it for your next planning task and notice whether the output is more useful than usual.
2. Try File Upload with one document you’ve been putting off reading. Most of us have something sitting in our downloads folder that we know we should have read by now. Upload it and ask Claude to pull out the parts most relevant to your role.
3. Test Custom Style with a piece of real writing. Find three or four paragraphs you’ve written that you’re pleased with, upload them, and then ask Claude to draft something new in that style. It won’t be perfect, but it will be a lot closer to your voice than the default output.

None of these features require any technical knowledge. They’re all built into the platform and available to explore at your own pace.

Six Claude Features Worth Knowing About (And Actually Using)
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

This tutorial covers how to use Google Gemini and Canva Code to generate interactive maths games from a simple text prompt. No coding, no subscriptions, no steep learning curve. Just a prompt, a few clicks, and a shareable link ready to go.

Watch this video to learn more:

What You Will Need

You have two routes here, and both are free:

• Google Gemini (gemini.google.com), with Canvas and Thinking mode enabled
• Canva Code (canva.com), using the Canva AI Code feature

Both work on the same principle: you type a description of the game you want, the AI writes the code, and you get a working, playable game. The whole process takes a few minutes.

Using Google Gemini to Build a Maths Game

Head to gemini.google.com and open the prompt box. Before you type anything, make sure two settings are active: Click on Tools and choose Canvas (Gemini’s coding platform) and Thinking mode. Enabling Thinking mode improves the quality of the output considerably.

Then type your prompt. Here is the one used for a number bonds game:

Hit submit. Gemini opens Google Canvas, generates the code, and builds the game. When you are happy with it, click Share to get a link you can post to your whiteboard, send to pupils, or share with colleagues.

Using Canva to Build a Maths Game

From the Canva homepage, click Canva AI, then select Code. Type your prompt into the box. Here is the one used for a fractions game:

Canva builds the game and gives you a Publish option. Add a title, publish, and copy the link. That link can go straight onto your interactive whiteboard or into a message to pupils.

Simply copy and paste these prompts into your AI tool of choice. Once entered, you can continue to refine your games by asking for edits.

Ready-to-Use Prompts

Here are four prompts you can copy and use straight away:

1. Number Bond Builder

Simple version:

Create an interactive maths game where students build number bonds to target numbers using number lines and ten frames.

Advanced version:

Create an interactive maths game where students build number bonds to target numbers using number lines and ten frames.

GAME FEATURES:
- Target number selector (5, 10, or 20)
- Two interactive modes: Number Line mode and Ten Frame mode
- Students select two numbers that add up to the target
- Visual representations show the addition happening
- Multiple rounds with different target numbers
- Score tracking and encouraging feedback
- "Check" button to verify the number bond
- "Next Challenge" button for new problems

NUMBER LINE MODE:
- Horizontal number line from 0 to target number
- Students click two points on the line to create a bond
- Visual arcs show the two parts combining
- Numbers displayed clearly at clicked points
- Equation appears below: ___ + ___ = [target]

TEN FRAME MODE:
- Classic 2x5 grid for bonds to 10, larger grids for 20
- Students click to fill squares with one color, then another color for the second number
- Visual grouping shows the two parts
- Equation updates as they click
- Option to use different representations (dots, counters, colors)

EDUCATIONAL FEATURES:
- Target numbers: bonds to 5, bonds to 10, bonds to 20
- Missing number challenges: 7 + ___ = 10
- All possible combinations shown as "collector card" achievements
- Commutative property demonstration (3+7 and 7+3 are the same)
- Speed round: find all bonds to 10 as quickly as possible
- Hints available showing remaining unfound bonds

DESIGN:
- Large, touch-friendly interface
- Clear visual distinction between the two addends
- Celebration animation when correct
- Running tally of discovered bonds
- Suitable for ages 5-8
- Clean, uncluttered layout

DIFFICULTY LEVELS:
- Easy: Bonds to 5 and 10
- Medium: Bonds to 20
- Challenge: Missing number problems

2. Fraction Pizza

Simple version:

Create an interactive visual fraction game where students match equivalent fractions using pizza slice representations.

Advanced version:

Create an interactive visual fraction game where students match equivalent fractions using pizza slice representations.

GAME FEATURES:
- Multiple circular "pizzas" showing different fraction divisions
- Drag and drop matching interface
- Students match equivalent fractions visually
- Pizzas show same total amount shaded but divided differently
- "Check Matches" button validates answers
- Visual overlay shows equivalent sections aligning
- Progress tracker showing matched sets
- Increasing difficulty levels

PIZZA REPRESENTATIONS:
- Each pizza is a circle divided into equal slices
- Shaded portions show the fraction
- Different pizzas show: halves, quarters, eighths, thirds, sixths, fifths, tenths

EQUIVALENT FRACTION SETS TO MATCH:
- 1/2 = 2/4 = 4/8
- 1/3 = 2/6
- 1/4 = 2/8
- 2/3 = 4/6
- 3/4 = 6/8
- 1/5 = 2/10
- 2/5 = 4/10

GAME MODES:
Mode 1 - Match Pairs: Drag pizzas together to show they're equivalent
Mode 2 - Sorting: Drag equivalent pizzas into groups
Mode 3 - Challenge: Given one fraction, build the equivalent using different denominator

EDUCATIONAL FEATURES:
- Fraction labels shown on each pizza (e.g., "3/4" and "6/8")
- Visual alignment when correct match made
- Number line appears showing all equivalent fractions at same point
- Simplification helper showing "simplest form"
- Interactive creator: divide your own pizza and see the fraction
- "Why are these equal?" explanation popup

VISUAL DESIGN:
- Realistic pizza appearance with toppings on shaded slices
- Clear division lines between slices
- Different colors for different fraction families
- Smooth drag-and-drop mechanics
- Celebration when matches confirmed
- Suitable for ages 8-11

FEEDBACK:
- Correct matches lock together with checkmark
- Incorrect matches bounce back with gentle shake
- Helpful hints: "Look at how much pizza is shaded"
- Progress bar showing sets completed

3. Shape Pattern Predictor

Simple Version:

Create an interactive pattern recognition game where students complete repeating patterns by dragging and dropping shapes.

Advanced version:

Create an interactive pattern recognition game where students complete repeating patterns by dragging and dropping shapes.

GAME FEATURES:
- Pattern sequence displayed with missing shapes (shown as empty boxes)
- Bank of shape options at bottom of screen
- Drag and drop interface to fill in missing shapes
- "Check Pattern" button validates answers
- Visual feedback for correct/incorrect placements
- Multiple difficulty levels
- Pattern rules shown after completion
- Score and streak tracking

PATTERN TYPES:
Simple AB patterns: circle, square, circle, square, ___, ___
ABC patterns: triangle, circle, square, triangle, circle, ___
AABB patterns: star, star, heart, heart, ___, ___
Growing patterns: small circle, medium circle, large circle, ___
Color + shape patterns: red square, blue triangle, red square, ___
Rotation patterns: arrow up, arrow right, arrow down, ___

DIFFICULTY LEVELS:
Level 1: Simple AB patterns, 1-2 missing shapes
Level 2: ABC and AABB patterns, 2-3 missing shapes
Level 3: Complex patterns, multiple missing shapes including middle gaps
Level 4: Growing/changing patterns
Challenge: Create your own pattern for a friend

EDUCATIONAL FEATURES:
- Pattern core highlighted when correct
- "What comes next?" prediction mode
- "What's missing?" mode with gaps in middle
- Pattern rule revealed: "The pattern is: triangle, circle, square, REPEAT"
- Option to extend pattern beyond the given sequence
- Hint button highlights the repeating unit
- Pattern creator mode for open-ended exploration

SHAPE LIBRARY:
- Basic shapes: circle, square, triangle, rectangle, pentagon, hexagon, star, heart
- Multiple colors: red, blue, green, yellow, orange, purple
- Size variations: small, medium, large
- Orientation: rotated versions of arrows/triangles

DESIGN:
- Clear boxes showing where shapes should go
- Drag-from-bank or click-to-select options
- Visual snapping when shape placed correctly
- Incorrect placements return to bank
- Running score display
- Suitable for ages 5-9
- Colorful, engaging graphics

PROGRESSION:
- Start with 5-shape patterns
- Increase to 8-10 shapes as difficulty rises
- Introduce multiple missing elements
- Add complexity with attributes (color AND shape)

4. Times Table Speed Challenge

Simple Version:

Create a timed multiplication practice game with progress tracking, adaptive difficulty, and motivating feedback. Player chooses which tables to practice: 2x, 3x, 4x ... up to 12x.

Advanced version:

Create a timed multiplication practice game with progress tracking, adaptive difficulty, and motivating feedback.

GAME FEATURES:
- Times table selector (choose which tables to practice: 2x, 3x, 4x... up to 12x)
- Timed rounds (30 seconds, 60 seconds, or 2 minutes)
- Questions appear one at a time with multiple choice or type-in answers
- Live score and timer display
- Streak counter for consecutive correct answers
- Progress bar showing questions completed
- Personal best tracking
- Star rating system based on accuracy and speed

GAME MODES:
Quick Fire: Answer as many as possible in 60 seconds
Survival: 3 lives, keep going until you lose them all
Target Practice: Specific times table focus (just 7x, just 8x, etc.)
Mixed Challenge: Random questions from all tables
Speed Climb: Questions get progressively harder

QUESTION PRESENTATION:
- Large, clear question display: "7 × 8 = ?"
- Multiple choice mode: 4 possible answers as buttons
- Type-in mode: Number pad or keyboard input
- Visual question (array of dots showing multiplication)
- Instant feedback: green checkmark or red x
- Correct answer shown if wrong

PROGRESS TRACKING:
- Session statistics: questions attempted, correct, accuracy %
- Times table mastery chart showing which tables are strongest/weakest
- Historical data: track improvement over time
- Badges/achievements: "Speed Demon" (10 in 30 secs), "Perfect Round", "Table Master"
- Personal records for each times table
- Weekly goals and streaks

EDUCATIONAL FEATURES:
- Hint system showing array representation
- "Show me how" button revealing counting/grouping strategy
- Mistake analysis: "You often struggle with 7 × 8, let's practice"
- Adaptive difficulty: focuses on weakest tables
- Commutative property reinforcement (7×8 = 8×7)
- Practice mode (untimed) vs Challenge mode (timed)

VISUAL DESIGN:
- Clean, distraction-free interface during play
- Bold numbers, easy to read quickly
- Traffic light colors: green (correct), red (incorrect), amber (time warning)
- Animated celebrations for streaks
- Progress graphs and charts on results screen
- Suitable for ages 7-11

MOTIVATION FEATURES:
- Countdown timer with visual progress
- Streak flames/lightning for consecutive correct
- Encouraging messages: "Great start!", "On fire!", "Almost there!"
- Level-up system (Bronze, Silver, Gold, Platinum)
- Comparison to previous attempts
- Optional: Gentle background music during play
- Share results feature

CUSTOMIZATION:
- Choose specific tables to include/exclude
- Adjust time limits
- Switch between multiple choice and type-in
- Enable/disable hints
- Sound effects on/off
- Choose difficulty: Beginner (up to 5×), Intermediate (up to 10×), Expert (up to 12×)

RESULTS SCREEN:
- Total correct / Total attempted
- Accuracy percentage
- Average time per question
- Tables breakdown (how many correct in each table)
- Stars earned (1-3 based on performance)
- "Try Again" and "Next Challenge" buttons
- Data saved to show improvement over time

Refining Your Game

Neither tool locks you in to a first attempt. If the game needs adjusting, go back to the prompt box (on the left of the screen in both tools) and ask for changes in plain language:

• “Change the colour scheme to blue and yellow.”
• “Add a 60-second timer.”
• “Make the numbers larger.”

The AI will update the game accordingly. It is a straightforward back-and-forth until you have something you are happy with.

Top Tip: You can always ask the Ai to improve your prompt for you. Paste in your basic prompt with a request to “improve this”

How to Use These Games in the Classroom

• Starter or plenary activity on the interactive whiteboard, giving pupils a few minutes of focused practice before or after a lesson.
• Independent practice at their own pace, using iPads or laptops with the shared link.
• Homework tasks, since the games are accessible from home via any browser, no login required.

Classroom Application: Three Steps to Get Started

1. Pick one prompt from the list above and try it in Gemini or Canva today. The whole process should take under five minutes. Having a working game in hand makes it much easier to decide whether this is worth building on.

2. Test it on your whiteboard before using it with a class. Check the link works from the board, that touch or click input behaves as expected, and that the game is set to an appropriate difficulty level.

3. Adapt the prompt to your current unit. Once you are comfortable with the process, write a prompt that matches exactly what your class is working on. If you are covering place value this week, build a place value game. The prompts are flexible, and the AI will follow your lead.

The games are not a replacement for good teaching, but they are a genuinely useful addition to the toolkit, and they cost nothing but a few minutes of your time to create

Make interactive Maths Games with AI: 4 Prompts To Use
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

This post pulls together some of the most important points from a webinar I run for trainee teachers and early career teachers, covering the things that can easily go wrong, and what you can do right now to get ahead of any problems.

Your Online Presence Is Already Professional

The moment you step into a school as a trainee, your social media life and your professional life start to overlap, whether you want them to or not. Parents will search for you. Students will search for you. In some cases, Ofsted inspectors conducting a school review will do a quick sweep of social media to get a sense of a school’s community. That includes staff.

Under the Teacher Standards, there is an expectation that teachers uphold public trust in the profession. Posts on social media are not considered private, even if that is how they feel when you are writing them. A legal adviser I came across put it plainly: anything you say that brings a school into disrepute can be used by your employer as grounds for dismissal.

That is not meant to be alarmist. It is just worth being clear about from the start.

Separate Your Personal and Professional Accounts

One of the most practical things you can do as a new or trainee teacher is to run two separate accounts: one private, one public.

Your private account is for friends and family. Lock it down. Use a username that is not easily searchable, approve followers manually, and be thoughtful about what you share there. Your public-facing account is where you follow educational communities, share resources, and engage with the wider teaching profession. That one can be open, because the content there is content you would be comfortable with anyone seeing.

It sounds like extra effort. In practice, it takes a few minutes to set up and saves a significant amount of worry later.

Your Digital Footprint Goes Back Further Than You Think

If you grew up with social media, there is a reasonable chance there are accounts, posts, and videos out there that you have completely forgotten about. An old YouTube channel. A Twitter account from your early teens. A Facebook profile you stopped using at university.

Do a search on your own name. Try Google, Facebook, and Twitter or X. See what comes up. You might be surprised. If you find old accounts you can no longer access, try to get back in and delete or disable them. If you can edit the information so it is harder to connect to you, do that.

This also applies to content you did not post yourself. Friends tagging you in photos, comments you are mentioned in, group posts you appeared in. You can turn off tagging permissions on most platforms, and it is worth doing.

One useful way to think about it: a digital footprint suggests something you leave behind that might fade. A digital tattoo is more accurate. Once content is shared, it is very difficult to fully remove.

Never Use Social Media to Contact Students

This one is not negotiable. Do not add students as friends or followers on any platform. Do not send direct messages. Do not accept follow requests from pupils, even if they are in sixth form and close in age.

If you need to contact a student or parent, use school email. That communication is stored centrally, can be reviewed if needed, and provides a clear record. Direct messages on social media leave no trail, and an absence of a trail raises questions you do not want to have to answer.

This comes up repeatedly in disciplinary tribunal records published by the DfE. It is not a rare edge case. Spend five minutes on that part of the DfE website and you will find case after case where a teacher’s career ended because of inappropriate contact made through social media.

What You Write in Private Can Still Become Public

A common mistake is assuming that a message sent in a private group chat, a WhatsApp thread with colleagues, or a locked account is genuinely private. It is not.

Anyone can take a screenshot. Anyone can forward a message. If you have typed something, it can be shared without your knowledge or consent. That comment about a class that felt like an offhand remark between colleagues can end up in a disciplinary hearing.

A straightforward test: before you post or send anything, ask yourself whether you would be comfortable with your headteacher reading it aloud. If the answer is no, do not send it.

Check Your Privacy Settings Regularly

Platforms change their settings frequently, and what was locked down last year may not be now. It is worth going into your accounts every few months and checking who can see your posts, who can tag you, and whether your friends list is visible to people outside your connections.

On Facebook, you can view your profile as it appears to someone who is not a friend. Use that. See what a parent or student would find if they searched for you today.

Turn off geotagging on your phone’s camera and within apps like Instagram. Photos taken at school during the day can carry location data that places you, and pupils, in a specific location. That is an unnecessary risk.

Passwords and Account Security

Use different passwords for different accounts, particularly for anything professional or linked to your school. A password manager such as LastPass or the one built into Google makes this manageable without needing to remember dozens of combinations.

Lock your devices when you step away from them. At school especially, an unlocked laptop or phone is a risk. Even if nothing malicious happens, if someone posts something from your account while you are away from your screen, it is almost impossible to prove it was not you.

Three Things to Do This Week

1. Run a search on yourself. Put your full name into Google, Facebook, and any platform you have used in the past. See what comes up and address anything you would not want a parent or employer to find.

2. Review your privacy settings. Go into each platform you use and check who can see your posts, who can tag you, and whether your location data is enabled. Make adjustments where needed.

3. Set up a separate professional account if you have not already. Even a simple public-facing profile where you follow educational communities gives you a space to engage professionally without mixing it with your personal life.

Book This Webinar for Your School or Training Provider

This session is available as a live webinar for initial teacher training providers, SCITT programmes, university education departments, and schools running their own ECT induction. It covers digital professionalism, online safety, and practical social media guidance for teachers at every stage of their career.

If you would like to book a session for your trainees or staff, get in touch via the WhiteboardBlog website and we can discuss dates and formats that work for your organisation.

Your Digital Footprint is Permanent – Here’s What Teachers Need to Know (Webinar)
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

Not as a future possibility. Not as a pilot project. It is already being used quietly, pragmatically, and increasingly routinely by teachers across the country.

Recent survey data from primary educators paints a consistent picture: most teachers are now using generative AI tools to support their work. They are planning lessons, creating resources, adapting materials for different learners, and developing their own subject knowledge. In primary science, some are beginning to use AI to design investigations and generate explanations for complex concepts.

The important point is this: teachers have moved ahead. Systems have not.

The question for school leaders is no longer “Should we allow AI?” It is: “How do we support staff to use it well?”

What the Data Tells Us

AI use is already embedded in day-to-day workflow. Most teachers have used AI tools within the past year, and many are now doing so weekly or more. Tools like ChatGPT dominate, but others are appearing alongside them.

The uses are practical, not gimmicky. Teachers are using AI to plan lessons and sequences, create worksheets and questions, adapt materials for different learners, and support their own subject knowledge. In science specifically, there is early use in designing investigations and generating explanations. AI is being used to improve core teaching work, not to replace it.

Workload is improving, but modestly. Most teachers report some reduction in workload. AI is helping with efficiency. It is not transforming teaching overnight, and that is a realistic and healthy starting point.

Teachers are acting professionally. Perhaps the most reassuring finding in the data is this: most teachers always check or edit AI output, and many have already identified inaccuracies or bias. Teachers are not blindly trusting AI. They are applying judgement. That matters, and it suggests the profession is responding responsibly.

Training and guidance are lacking. A large proportion of teachers report no formal training, limited or self-taught use, and unclear or absent school policies. So we now have a situation where AI use is widespread, confidence is mixed, and support is inconsistent. This is where leadership matters most.

The Risk of Doing Nothing

It can be tempting to take a “wait and see” approach, assuming the technology will settle before schools make decisions. The problem is that staff are not waiting.

Without clear expectations, practical guidance, and professional development, teachers will continue to use AI independently, develop habits in isolation, and rely on trial and error. That creates inconsistency at best.

Doing nothing is not a neutral position. It simply leaves staff to work it out alone.

What Schools Should Do Now

Set clear, proportionate expectations. Start by answering a straightforward question: what do we want AI use to look like in our school? This does not need to be a lengthy policy document. Shorter is often better. Focus on acceptable uses (planning, drafting, resource creation), non-negotiables (checking accuracy, protecting pupil data), and professional responsibility. Avoid blanket bans or overly restrictive rules that do not reflect the reality of what is already happening. A useful working principle is this: AI can support professional judgement, but it does not replace it.

Prioritise practical CPD. The survey is clear that teachers want help, but it needs to be practical. Focus professional development on how to write effective prompts, how to refine and improve outputs, how to check for accuracy and bias, and how to use AI for real classroom tasks. Avoid long theoretical sessions about how AI works or generic discussions about the future of education. Teachers need to see how to use this for tomorrow’s lesson. We can help with that. Get in touch!

Make it subject-specific. One of the strongest signals from the data is that teachers want subject-relevant support. In primary science, this could include generating investigation ideas, identifying and addressing misconceptions, creating age-appropriate explanations, and designing retrieval questions. Without subject-specific guidance, there is a risk that AI use becomes superficial or detached from curriculum intent. Done well, it can genuinely strengthen subject knowledge and planning.

Build a culture of critical use. Teachers in the survey are already checking outputs. Schools should reinforce this by making it explicit: always verify AI-generated content, compare it with trusted sources, and be alert to bias and simplification. This is not just about staff. It is also about modelling good practice for pupils, particularly in science, where evaluating evidence is central to what we teach. AI provides a useful opportunity to strengthen critical thinking, not weaken it.

Support the hesitant, not just the confident. Not all staff are using AI with confidence. Where teachers are not engaging, the reasons are predictable: lack of confidence, concerns about accuracy, and uncertainty about what is expected. This is not resistance; it is caution. Schools should provide low-stakes opportunities to explore, share simple starting points, and avoid assuming prior knowledge. A small amount of structured support can make a significant difference.

Classroom Application: Three Steps to Take Now

1. Start the conversation with staff. Before writing a policy, find out what your staff are already doing. A simple anonymous survey or a short staff meeting discussion will give you a much clearer picture than assumptions will. You may find that AI use is already more embedded than you realised.

2. Identify one subject area to pilot subject-specific guidance. Rather than trying to produce whole-school AI guidance in one go, pick a subject where you have a confident lead and develop practical guidance there first. Primary science is a good starting point given the evidence around investigation design and misconception-checking.

3. Run a short, practical session rather than a lecture. Give staff 20 minutes to try a single AI task relevant to their planning, then discuss what they noticed. A brief, hands-on experience will do more than an hour of slides about what AI is and where it came from.

A Leadership Moment

What stands out most from this survey data is not the technology itself. It is the professionalism of teachers. They are adopting new tools, applying judgement, managing risk, and looking for ways to improve their practice.

The role of school leadership is to match that professionalism with clarity, support, and direction.

AI is not coming into schools. It is already there, being used every day by teachers trying to do their job well. The question is no longer whether to engage with it. It is whether schools will provide the support needed to ensure it is used safely, thoughtfully, and effectively.

Get that right, and AI becomes what it should be: a tool that supports good teaching, not something that sits outside it.

AI in Primary Schools: What Leaders Need to Do Next
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

Which is both true and completely useless, because every other applicant writes exactly the same thing.

The problem is not that you have nothing to say. It is that you are too close to your own experience to see what is actually valuable in it. And most of us have never had to translate teaching practice into formal written language before.

This is where AI tools can help, and it is worth being clear from the outset about how. AI is not a ghostwriter. If AI writes your personal statement, you will not be able to talk about it fluently at interview. Schools are looking for you, not a polished output. What AI can do is act as a thinking partner that helps you see what you already have to offer.

Getting Past the Blank Page

Before you touch the application form, open a chat with Claude or ChatGPT and do a brain dump. Just type out everything you can think of: lessons that went well, moments that did not, students you found difficult and how you handled it, feedback from mentors, things you tried that worked, things that failed and what you learned from them. Do not edit it. Do not structure it. Just get it out.

Then give the AI a prompt along these lines:

What comes back is a reflection of your own experience, organised in a way that is hard to see when you are inside it. The AI might notice that you mention building structured routines for pupils with additional needs in three different examples, suggesting it is a genuine strength. That is useful information.

Here is where most people go wrong: they take what the AI produces and paste it straight into the application. Do not do this. The AI version sounds like an AI wrote it. It is smooth, well-structured, and sounds like absolutely nobody in particular. Headteachers read hundreds of these. Take the themes the AI has identified and write your own response to them, using your specific examples and the language you would actually use with a colleague.

Using AI to Strengthen a Draft

Once you have a first draft, AI becomes something different: an editor. Good feedback on applications is hard to come by. Your mentor is busy, your tutor may not look at it until the week it is due. But you need someone to tell you where your draft is weak before you submit it.

The key is asking specific questions rather than something vague like “is this good?” Try these instead:

• “Where in this statement am I making vague claims that are not backed up with a specific example?”
• “What questions would a headteacher still have after reading this?”
• “Read this as a sceptical interviewer. What would you push back on?”

That third question is particularly worth doing. It forces the AI to stress-test your claims rather than validate them. If you write “I am skilled at adapting lessons for different learning needs,” a good prompt will push back: what is your evidence? When did you actually do that, and what happened?

One word of caution: AI feedback is generally reliable at identifying vagueness and structural weaknesses, but less reliable at judging whether your specific examples are actually impressive, because it does not know your context. If it tells you something reads as weak and you believe it was genuinely significant, trust your own judgement.

Covering Letter and CV: Quick Checks

For your covering letter, ask the AI to check its structure. Does it open with something specific to this school, or does it sound like a template? Does it end with a clear statement of intent, or does it trail off? These are structural questions that take seconds to check.

For your CV, ask it to read for gaps. Are there unexplained periods? Is anything ambiguous about your subject specialism or placement experience? Better to find those now than during scrutiny.

Interview Preparation

This is where AI can be most immediately useful, and where trainees are least likely to use it. Take the job description and person specification for the role you are applying for, paste both into the chat, and ask something like this:

What you get back is a tailored question bank that reflects the specific priorities of that school. A school that mentions inclusion fifteen times in its documentation is going to ask about inclusion. The AI will notice that.

Once you have the questions, practise by speaking, not by typing. Say your answer out loud, then type up what you actually said, and ask for feedback. The gap between what you think you are going to say and what actually comes out when you open your mouth: that is the thing to close before the day.

You can also ask the AI to role-play as a sceptical panel member and challenge your answers. If you say “I build strong relationships with students,” it should ask: how, specifically? Can you give an example where that relationship was tested?

Do Not Overlook the Observation Debrief

If you are asked to teach a lesson as part of the interview day, you will almost certainly be asked to evaluate it immediately afterwards. Questions like: what went well, what would you change, how did you know the students were learning? Give the AI a brief description of a recent lesson you have taught and ask it to generate likely debrief questions. Then practise your answers.

Questions to Ask the Panel

Candidates spend hours preparing answers and about thirty seconds preparing questions to ask. It shows. Ask the AI something like: “Given this person specification, what would be good questions for me to ask the panel that show I have thought seriously about this role?” You will end up with questions that are genuinely thoughtful, rather than asking whether the school has a good CPD offer, which is what almost everyone asks.

Three Things to Try This Week

• Brain dump first. Before you open the application form, spend twenty minutes typing your placement experiences into an AI chat without editing. Ask it to reflect back the two or three strongest themes. Then write your statement from those themes in your own words.
• Stress-test your draft. Paste your personal statement into a chat and ask the three specific editing questions listed above. Use the feedback to strengthen your examples, not to replace your voice.
• Build a question bank from the job spec. Paste in the job description and person specification and ask for a tailored set of ten interview questions. Practise your answers out loud.

A Final Thought

AI does not get you the job. You do. What it can do is help you get out of your own way: organising the experience you already have, giving you feedback when there is nobody else to ask, and helping you practise at eleven at night when your mentor is not available.

The schools you are applying to want to see your thinking, your values, and your specific experience. AI is just the tool that helps you get that onto the page.

Good luck. The fact that you are thinking carefully about how to present yourself is already a good sign.

How AI Can Help You Prepare for Your First Teaching Job Application
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

In my experience of working in EdTech for over twenty five years, this is one of the most common and least discussed patterns in technology purchasing. It is worth saying clearly: it is not usually caused by bad tools. More often, it is caused by buying something before the school has a clear answer to what problem it is actually trying to solve.

These five questions will not prevent every poor purchasing decision, but they will help you go into the process with your eyes open.

1. What specific problem are we trying to solve?

Not “how can we use technology better?” or “what are other schools using?” A specific, named problem.

If a school cannot articulate the problem in a sentence or two, the purchasing decision will be similarly vague. EdTech demonstrations are very good at creating a sense of possibility, but possibility is not the same as need. The question that most demos never ask is the one you should ask first: what, precisely, is broken, slow, or difficult right now, and is a piece of software a reasonable way to address it?

If you cannot answer that clearly before the demo, it is worth pausing rather than continuing.

2. Do we actually have that problem?

This sounds like a repeat of the first question, but it is a little different.

The EdTech market is skilled at helping schools identify problems they did not know they had, particularly when the available solution happens to address them. A persuasive demonstration can make a fairly ordinary challenge feel urgent, or suggest that current practice is more inadequate than it actually is.

It is worth separating the question of what a tool does from the question of whether that genuinely reflects something your school is struggling with. Bring in a colleague who was not at the demo and describe the problem without mentioning the tool. See if they recognise it.

3. What will change about how teachers work?

Every tool that actually sticks requires a sustained change in practice. Not a one-off training session, but a real shift in how someone does part of their job, probably week in, week out.

If there is no clear answer to what that change looks like in practice, or whether teachers have the time and capacity to make it, the tool is unlikely to survive contact with a busy spring term. This is worth discussing honestly before any commitment is made, with the people whose practice would actually need to change.

4. What does it actually cost?

The licence fee is only part of the answer.

Add setup time, training time, ongoing support, and the staff time involved in building it into existing workflows. EdTech is regularly undercosted at the point of purchase because the monthly fee looks reasonable and the hidden costs remain invisible until someone tries to implement the thing properly.

A tool that saves ten minutes of teacher time per week and requires three days of staff time to embed and maintain is not a saving. It is worth doing a realistic, fully costed estimate before committing, including what it would take to exit the contract if the tool does not work as expected.

5. What does success look like, and who will check?

This is the question that schools most consistently skip.

If nobody is responsible for evaluating whether the tool is achieving anything, nobody will notice when it gradually stops being used. A simple criterion agreed at the point of purchase, something like “in six months, this is what should look different,” together with a named person who will check on it, is worth more than any amount of usage data from the vendor’s dashboard.

The absence of an evaluation plan is not just an oversight. It is one of the main reasons subscriptions run on quietly into their third year without anyone being sure whether they are still useful.

Three Practical Steps to Move Forward

Before agreeing to any new EdTech purchase, try these three steps.

First, write the problem down before you book the demo. A single sentence describing what the tool needs to fix, agreed by more than one person. If you cannot agree on the sentence, you are not ready to buy anything.

Second, cost it properly. Add up the licence fee, the setup time, and a realistic estimate of ongoing staff time. Compare that to the estimated benefit. If the numbers do not add up comfortably, the tool needs to justify itself more clearly.

Third, set a review date and name a reviewer. Agree before purchase on a specific date six months in and a specific person who will check whether the tool is still being used and whether anything has actually changed as a result.

One More Thing

Most schools are currently paying for at least one tool that nobody is actively using. Before making any new purchasing decision, a thirty-minute audit of active subscriptions is usually time well spent. The five questions above apply equally to what you already have, not just to what you are considering next.

For more on EdTech tools that are worth the investment for primary schools, there is plenty more on this blog. And if any of this is relevant to a wider school or MAT context, details of training and consultancy are available here.

Before You Buy the Next EdTech Tool: Five Questions Worth Asking First
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

Why It’s Worth Ten Minutes of Your Time

Before we get into the prompts themselves, it’s worth being honest about where AI is actually useful for teachers, because it isn’t useful for everything.

The tasks where it genuinely earns its place are the ones that eat your planning time without requiring much professional judgement: writing a first draft, generating a set of questions, simplifying a chunk of text. These are tasks where the blank page is the problem, not the thinking behind it.

Writing a set of differentiated questions on a topic used to take the better part of 40 minutes. With AI, the same task takes around three. The questions aren’t perfect straight away, but they’re most of the way there, which means the work left is editing rather than starting from scratch. A sensitive parent email that might take 20 minutes of careful drafting can have a solid first version in about 30 seconds. A revision resource for yesterday’s lesson, something students can actually use, takes roughly four minutes to produce. Those are the kinds of tasks teachers find themselves reaching for AI to help with most.

The Tool to Start With: Google Gemini

If you search “AI tools for teachers” right now, you’ll find about 40 different recommendations and a fair amount of debate about which is best. Ignore all of that for now.

There is one tool worth starting with, and the reason for recommending it is straightforward: you probably already have access to it. It’s called Google Gemini. If you have a Google account, you have access. Many schools run on Google, so it may be available through your school account. If you have a personal Gmail, that works too. There is nothing to download, nothing new to set up. Go to gemini.google.com, sign in, and you will see a large text box. Type in what you want it to do, press enter, and it responds.

That really is the basic idea. It is a conversation. You type a request, it replies, you can reply to that reply, and the process continues until you have something useful.

Prompt One: Multiple Choice Questions

Type something like this into Gemini:

“Write five multiple choice questions about the water cycle for Year 7 students.”

In around 15 seconds you will have five questions with answers and a brief explanation for each. Check them through, because AI does occasionally get things wrong or produce questions that are slightly off for your context, but the blank page has been filled. You have something to work with.

You can change the topic, the year group, the number of questions, or the format. Ask for short answer questions instead. Ask for ten questions rather than five. Ask it to focus on a specific aspect of the topic. Once you have run through this once, adapting the prompt takes about five seconds.

When you want to move the content elsewhere, there is a copy button at the bottom of the response. Paste it into Word or whichever document format you work in.

Prompt Two: Simplified Text

Go to a source of information on your topic, An online textbook you have access to, or maybe try BBC Bitesize, and copy a chunk of text. Then paste it into Gemini with an instruction like:

“Rewrite this for a student who finds reading difficult. Keep the key information, but use simpler language and shorter sentences.”

You can specify the reading level, the year group, or any other requirements. The result won’t always be perfect, but it gives you a starting point that would have taken considerably longer to produce by hand. Edit what needs editing, and you have a differentiated resource in a fraction of the usual time.

Prompt Three: The Parent Email

This is the one that teachers tend to share with colleagues.

“Write a polite but firm email to a parent explaining that their child has been struggling to submit homework on time. The tone should be professional and non-confrontational. The student’s name is Jamie.”

You will get a draft back within seconds. Read it through, personalise the details, and adjust anything that doesn’t sound right for your school’s context. If the tone isn’t quite right, you can ask Gemini to adjust it: “make it slightly warmer” or “add a line about wanting to work together.” It will revise accordingly. The sentences you might have spent ten minutes agonising over are already written. You are refining rather than starting from scratch.

Two Things That Trip People Up

Vague prompts produce vague results. If you type something like “write me a lesson plan,” you will get something generic that probably isn’t much use, and you might conclude that AI isn’t worth the effort. The tool isn’t the problem. The more specific you are about year group, subject, time available, and what you actually need included, the more useful the response will be. Compare “write a lesson plan” with “write a 50-minute lesson plan for Year 9 science on the structure of DNA, including a starter activity, a main task with differentiation for lower ability students, and an exit ticket.” Same tool, very different output. This guide covers better prompting techniques.

The first response isn’t the final version. Don’t hand in the first draft of a report you’ve written, and the same principle applies here. Ask it to make the response shorter, change the tone, add something specific, remove something that isn’t right. The back and forth is where most of the value actually comes from.

Classroom Application: Three Steps to Try This Week

1. Start with one task, not several. Pick one thing on your to-do list this week, a worksheet, a set of questions, a letter home, a revision summary, and ask Gemini to do it. One task, done. See how it goes before adding anything else to the process.

2. Practise adjusting your prompts. Once you have a first response, ask Gemini to change something about it. Make it shorter. Change the reading level. Adjust the tone. Getting comfortable with this back-and-forth is what makes AI genuinely useful rather than occasionally helpful.

3. Check everything before it goes anywhere. AI makes mistakes. Questions can be inaccurate, emails can miss the right tone, simplified text can lose important nuance. Build in a quick read-through as a habit before anything goes to students or parents.

A Reasonable Starting Point

AI will not transform your workload overnight, and it is worth keeping expectations realistic. What it does is reduce the time spent on specific, repeatable tasks so that you can focus on the parts of the job that require your professional judgement.

Three prompts. One tool. That is a reasonable place to begin.

Beginners Guide Videos

For a useful set of tutorials that will get you started with AI, I’ve put together this playlist to guide you.

Your First Three AI Prompts as a Teacher (Getting Started with AI)
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

Google Gemini has a feature designed to fix this. It is called Gems, and it is worth knowing about.

For example, here’s a gem that generates science activities you could use in the classroom. Just tell it what topic you’re teaching.

What is a Gem?

A Gem is essentially a saved AI assistant. You write the instructions once, attach any files you want it to reference, give it a name and a brief description, and save it. From that point on, you can start a fresh conversation with that assistant whenever you need it, without having to set anything up again.

Think of it as a version of Gemini that has already been briefed. It knows what you want it to do, it has the context it needs, and it is ready to go.

Gems are available on Google Gemini’s main page at gemini.google.com. Google also provides a small selection of pre-made Gems to get you started, but the real value comes from building your own.

This video explains more:

Why This is Useful for Teachers

The most obvious use case is saving time on repetitive tasks, but the bigger opportunity is building something that genuinely fits your curriculum and your school context.

In the video above, the example used is a Science Activity Generator. The Gem was given a system prompt explaining that it should behave as a skilled science teacher and suggest creative activity ideas appropriate for a given year group and topic. It was also given two text files: one listing different presentation methods (poem, role play, game, song and so on) and one mapping content to the primary National Curriculum. A PDF of the full curriculum was included too.

The result is an assistant that, when you type “Year 4, digestion”, returns five structured lesson activity ideas tailored to that year group. You can then ask it to expand on any of those ideas in more detail. All of that happens without you having to paste in your instructions or upload your files again. They are already there.

This kind of Gem could be useful for:

• Generating lesson activity ideas across topics and year groups
• Creating differentiated task descriptions from a unit overview
• Summarising research or reading for a specific key stage
• Producing parent-facing explanations of curriculum content
• Supporting planning for supply cover or SEND adaptations

How to Create a Gem: Step by Step

Step 1: Go to Gems

From the Gemini home page, look for the Gems option in the left-hand menu and click on it. You will see any existing Gems, including the pre-made ones from Google. To create a new one, click “New Gem.”

Step 2: Give it a name and description

Choose a clear, practical name: “Science Activity Generator,” “Year 6 Writing Prompt Builder,” “Assessment Question Creator.” Add a short description explaining what it does. This becomes useful when you share it with colleagues, as it tells them what to expect and how to use it.

Step 3: Write your instructions

This is the system prompt: the instructions that tell Gemini how to behave. Be specific about what you want it to do, what format you want responses in, and any constraints it should follow. It helps to prepare this in a document first so you can paste it in cleanly.

If you are not sure whether your prompt is well-written, Gemini has a built-in option to suggest improvements to your instructions. It is worth using if you are new to prompt writing.

Step 4: Add your files

Click “Add Files” to upload any documents you want the Gem to reference. These could be curriculum documents, unit plans, reading lists, example texts, or your own resource collections. Supported formats include text files and PDFs.

Step 5: Test it

Before saving, run a test conversation to check the Gem is behaving as expected. Ask it something representative of how you plan to use it. If the output is not quite right, adjust your instructions and try again.

Step 6: Save and use it

Once you are happy, click Save. The Gem will appear in your Gems list. Each time you open it, you start a fresh conversation with the same configured assistant.

Sharing Gems With Colleagues

This is where Gems become particularly useful at a department or school level. Once you have built a well-configured Gem, you can share it with colleagues so they do not have to build their own version from scratch.

To share a Gem, click the sharing icon and either type in the email addresses of specific colleagues, or change the access setting so that anyone with the link can use it. Copy the link and share it however works best in your school, whether that is via email, a shared drive, or a staff intranet.

This makes Gems a practical way to build a shared set of AI tools that are consistent, contextualised to your school, and ready for anyone on the team to use.

Classroom Application

Here are three practical starting points you could try this week:

1. Build a subject-specific activity generator. Pick one subject you teach regularly. Write a prompt that tells Gemini to suggest activity formats for a given topic and year group. Add your curriculum overview or unit plan as a reference file. Test it with two or three topics and refine the instructions until the output is useful.
2. Create a shared planning aid for your team. If you are a subject lead or part of a planning team, build a Gem that reflects your school’s curriculum context and share the link with colleagues. This gives everyone access to the same starting point without duplicating effort.
3. Use Gems to reduce repetitive admin tasks. Think about the prompts you find yourself typing repeatedly: writing parent updates from notes, generating quiz questions from a topic list, summarising pupil feedback. Turn one of those into a Gem and save yourself the setup each time.

A Note on Getting Started

If you are new to writing system prompts, the simplest approach is to start with what you already do manually and describe it clearly. “You are a KS2 science teacher. When I give you a topic and year group, suggest five creative ways to teach it” is a reasonable starting point. You can always refine it once you see how it responds.

Gems are available to users with a Google account and access to Gemini. Some features may depend on your subscription level or whether your school uses Google Workspace.

The video above walks through the full process with a worked example if you want to see it in action before building your own.

Create and Share Google Gems: Make a Library of AI Assistants
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

What Are Sketch Notes?

Sketch notes are visual summaries that combine text and simple illustrations to capture ideas in a memorable way. They’re popular in education because they support visual learners, make abstract concepts more concrete, and give pupils something to refer back to rather than a wall of typed text.

The good news here isn’t that AI has suddenly become creative. It’s more practical than that: Gemini can do a reasonable first draft of a sketch note, quickly, which you can then refine or use as a starting point.

What You’ll Need

Just a Google account and access to gemini.google.com. No special subscription is required for the basic image generation feature, though availability may vary.

Step 1: Start a Chat in Gemini

Go to gemini.google.com and open a new chat. Before you do anything else, check the mode setting. Switch from Fast to Thinking for a better quality response. It takes a little longer, but the output is noticeably more considered.

Step 2: Ask Gemini to Explain Your Concept

Start with a clear explanation prompt. For example:

“Can you explain why we get seasons and why the earth’s tilt is important? Please explain this for 10-year-olds.”

Gemini will generate a written explanation. Don’t skip this step. The explanation gives Gemini the content it needs to produce a meaningful visual rather than a generic one.

Step 3: Ask for a Sketch Note

Once you have the explanation, follow up with a second prompt asking Gemini to turn it into a sketch note. A prompt along these lines works well:

“Using your visual thinking skills, create a hand-drawn sketch note visual summary of this explanation. Use a sketchy pen style, suitable for a 10-year-old.”

Then, importantly, click on the Tools option below the text box and select Create Image before hitting submit. This tells Gemini you want a visual output, not more text.

Step 4: Review and Refine

Gemini will generate your sketch note. Click on it to view it at full size, or hover over it and select the download option to save a full-size version.

Check it carefully. You may spot errors, odd word choices, or details that don’t quite work. If so, ask Gemini to regenerate. You can be specific: “Please redo this without the word ‘flashlight’, using ‘torch’ instead”, for example. A couple of iterations usually gets you something usable.

Going Further: More Complex Topics

The same approach works for more abstract or curriculum-heavy topics. Try something like:

“Describe AI for beginners. Explain how it can be used in education.”

Generate the explanation first, then use the sketch note prompt with the Create Image tool. Gemini will produce a more complex visual summary. It won’t always be perfect, but it gives you a solid starting point.

Classroom Application

Here are three ways to use this in your teaching straight away:

1. Pre-lesson visual primers. Generate a sketch note for the key concept of your next lesson and display it on the board as pupils arrive. It works as a low-key retrieval cue or a preview of what’s coming.

2. Pupil-facing revision resources. Produce sketch notes for revision topics and share them digitally or print them for books. They’re particularly useful for pupils who find dense written notes hard to process.

3. Modelling the skill. Show pupils a Gemini-generated sketch note, then ask them to create their own by hand on the same topic. It gives them a model to work from and makes the task feel more achievable.

A Note on Accuracy

AI-generated images, including sketch notes, can contain errors. Words may be misspelled, details may be inaccurate, and occasionally the visual logic won’t quite make sense. Always review the output before sharing it with pupils. It’s worth treating Gemini’s sketch notes as a draft rather than a finished resource.

Worth Experimenting With

The sketch note prompt is something you’ll want to play around with. Adjusting the age range, the style description, or the complexity of the topic will give you different results. There’s no single perfect formula; it depends on the concept and what you need the visual to do.

Give it a try with a topic you’re teaching next week. It takes about five minutes from start to finish, and even an imperfect output can save time compared to building a visual from scratch.

How to Create Sketch Notes in Seconds with Google Gemini
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

Gemini Canvas changes that. It’s a feature built into Google Gemini that lets you describe what you want and generates a working, interactive game for you. No code. No specialist knowledge. Just a prompt and a few minutes.

Here’s an example game I made.

Here’s how it works.

What Is Gemini Canvas?

Canvas is a mode within Google Gemini that lets you generate and edit interactive content, including HTML-based games and tools. When you switch to Canvas mode, Gemini doesn’t just give you a text response: it builds something you can actually use and share.

You access it at gemini.google.com. Once you’re there, click on Tools and select Canvas. It’s also worth turning on Thinking mode, which gives Gemini a little extra processing time to produce better results.

Building a Game: A Step-by-Step Walkthrough

To show how this works in practice, here’s an example using animal classification, a popular topic across KS2 science.

Step 1: Write Your Prompt

The prompt doesn’t need to be complicated. Something like this is enough:

“Create an interactive game for primary students about animal classification. Include categories for mammals, fish, amphibians, reptiles, and birds. Students should be able to click a button to see a random vertebrate animal, and then click the correct category to sort it.”

That’s it. Submit the prompt with Canvas selected and Gemini will start building.

Step 2: Test and Refine

Within a minute or so, you’ll see a preview of the game on screen. Try it out. Click through a few animals, check the categories work, and see what you think.

In the example from the video, the first version worked well but only included 15 animals, which would limit replayability. A follow-up prompt asking for 30 animals, including some trickier ones, updated the game automatically. Another prompt requested that animals appear in a randomised order with no repeats until the full list had been shown. Again, Gemini updated the game straight away.

This is where Canvas is genuinely useful. You build iteratively. Start with a basic version, then refine it through conversation until it does what you need. You’re not locked into your first attempt.

Step 3: Share With Your Class

Once you’re happy with the game, click the Share button at the top of the screen. This generates a link you can post on your VLE, share via email, or turn into a QR code for pupils to scan with iPads or tablets.

You can also copy the underlying code and share it with colleagues. They can use it as a starting point for their own versions, adapting it for a different topic or year group.

What Kinds of Games Can You Make?

The animal classification example is just one possibility. The same approach works for a wide range of curriculum topics. You could ask Gemini to build:

• A times tables practice game with a score tracker
• A vocabulary matching activity for a history unit
• A true/false quiz on states of matter
• A spelling game that reads out words for pupils to identify

The prompts don’t have to be long or technical. Describe what you want the children to do, what subject it’s for, and roughly how it should work, and Canvas will produce a working first draft.

A Note on Limitations

Like any AI tool, Canvas won’t always get it right first time. You may need to tweak the prompt or ask for adjustments. Occasionally the generated game has a bug or behaves oddly; in those cases, describe the problem and ask Gemini to fix it.

It’s also worth previewing any game before you use it with a class, both to check it works and to make sure the content is accurate. AI-generated content is a starting point, not a finished product. A quick review before you share takes very little time and avoids awkward moments mid-lesson.

Classroom Application

Three things to try this week:

1. Pick one upcoming topic and build a sorting or classification game. Animal classification, properties of materials, and parts of speech all lend themselves well to this format. Write a simple prompt describing what you want and see what Gemini produces.

2. Refine through conversation. Don’t expect the first version to be perfect. Ask Gemini to add more items, change how the scoring works, or adjust the difficulty. Each prompt updates the game in real time.

3. Share via QR code. Once your game is ready, generate a shareable link and convert it to a QR code (there are free tools online for this). Pupils can scan and play on their devices without you needing to type out a long URL or set anything up in advance.

Worth Knowing

You don’t need a premium Gemini subscription to use Canvas; the basic version is free. The games are web-based, so pupils just need a browser to access them. No app downloads or logins required on their end.

If you share the code with a colleague, they can paste it into Canvas and continue editing from there, which makes it a useful way to collaborate on resources across a year group or key stage.

Building interactive resources used to mean either buying something off the shelf or spending a long time learning a tool from scratch. Canvas sits somewhere more useful: it handles the technical side, and you handle the teaching logic. That’s a reasonable division of labour.

Make Your Own Classroom Games in Minutes with Google Gemini Canvas
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

Custom GPTs are a way around this. Rather than retraining ChatGPT at the start of every session, you build a version of it that already knows what it needs to know, follows the instructions you have set, and has access to the documents you have uploaded. Once it is set up, you just open it and start asking questions.

This tutorial walks through the process of building one from scratch, using a science curriculum assistant as the example.

Google Gemini does a very similar thing with Gems. If you want to know how to do it in Gemini, watch this video.

What is a Custom GPT?

A custom GPT is a version of ChatGPT that you configure yourself. You give it a set of instructions that define its role and behaviour, and you can upload files such as PDFs that it will use as a knowledge base. When you or a colleague opens that GPT and asks a question, it responds within the boundaries you have set.

You can think of it like setting up a specialist colleague who only knows about the things you have briefed them on. If you upload the National Curriculum for Science, your GPT will answer questions about that curriculum. If you also upload your school’s assessment grids or a knowledge matrix, it can draw on those too.

Creating and saving custom GPTs requires a ChatGPT Plus subscription, which costs around £20 per month. Once they are built, however, you can share them with colleagues via a link, so other staff members can use them without needing their own Plus account. Here’s a link to the science one I create in the video.

This tutorial shows how to build a Science assistant, but just swap the documents for your own subject curriculum guides and it’ll work for any subject.

Building a Science Curriculum Assistant: Step by Step

Step 1: Write your instruction prompt

Before uploading anything, you need to tell the GPT what it is for. This is the instruction prompt, and it is the most important part of the setup.

A useful prompt for a science curriculum assistant might include:

• Its role: an expert in the National Curriculum for Science.
• What it should do: identify curriculum objectives, suggest lesson activities, clarify statutory and non-statutory content.
• How it should respond: clear headings, bullet points, warm and supportive tone.
• What it should avoid: heavy worksheet use, overly academic language.

Prepare this in a text editor before you start, so you can paste it in directly when prompted.

Step 2: Create the GPT

In ChatGPT, click on Explore GPTs in the left-hand menu, then click Create. You will be taken into a setup conversation where ChatGPT asks you questions about what you want your GPT to do.

Paste in your instruction prompt when asked. ChatGPT will suggest a name and generate a profile picture automatically. You can accept these or adjust them. It will also ask follow-up questions about tone and emphasis, which gives you a chance to refine things without having to rewrite your prompt manually.

Step 3: Upload your documents

Switch to the Configure tab. This is where you can upload the files that form your GPT’s knowledge base. For a science curriculum assistant, you might upload:

• The National Curriculum for Science (PDF).
• Your school’s assessment grids or knowledge matrices.
• Any planning templates or progression documents you regularly refer to.

Click Upload files and select each document. The GPT will be able to search and reference these whenever it responds to a question.

Step 4: Test it before saving

Use the preview panel to try a few questions before you finalise the GPT. In the video, the example prompt is asking for Year 1 statutory objectives for animals, including common misconceptions and suggested practical activities. Try questions that reflect how you would actually use the tool day to day.

If the responses are not quite right, you can either ask the GPT to adjust its approach in the conversation, or go back to the Configure tab and update the instructions directly. It is worth spending a few minutes on this before saving.

Step 5: Save and share

When you are happy with how it is working, click Create. You will be asked whether to keep it private, share it with anyone who has the link, or publish it publicly in the GPT store.

For most school use, the link option is the most practical: you can email the link to colleagues and they can access the GPT directly, without needing to build it themselves.

Classroom Application

Three ways to put this into practice straight away:

1. Build a curriculum assistant for your phase. Upload the relevant National Curriculum sections and your school’s assessment documents. Use it to quickly pull out statutory objectives, check curriculum coverage, or draft activity ideas for a topic you are about to teach.
2. Create a marking and feedback GPT. Upload your school’s mark scheme and a description of the year group and task. Use it to help draft written feedback more quickly, keeping the language consistent with your school’s expectations.
3. Share it with your team. Once built, send the link to colleagues. A single well-configured GPT can save planning time across a whole year group or department, without everyone needing their own subscription.

A Note on Cost and Free Alternatives

Custom GPTs require a ChatGPT Plus subscription. If you want to try the feature without committing to an ongoing subscription, one option is to sign up for a single month, set up a handful of GPTs, and then cancel. The GPTs you have created will still be accessible via their links.

If you would prefer a free option, Google Gemini offers a very similar feature called Gems. The setup process is comparable and the core functionality is much the same. There is a separate tutorial on Gems on the channel if you want to compare the two before deciding which to use.

Summary

Custom GPTs are a practical way to reduce the setup time involved in using ChatGPT regularly. Rather than re-explaining your context every session, you build a version that already has that context built in, with your documents uploaded and your preferences set. For science teachers in particular, having a curriculum assistant on hand that knows your documents and responds in a style that suits you is a reasonable time-saver, especially as a starting point for planning and assessment conversations.

If you build one, the video includes the full step-by-step walkthrough.

Let us know in the comments how you get on, or if there are other subject areas you would like to see covered.

How to Create a Custom AI Assistant in ChatGPT: A Step-by-Step Guide for Teachers
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

This playlist is an attempt to fill that gap. It is a collection of short tutorial videos covering some of the most useful AI tools available to teachers right now. Each one is focused on doing something specific and practical, rather than giving a broad overview of what AI might do in the future.

You do not need any previous experience with AI tools to follow along. If you have got ten or fifteen minutes and a browser open, that is enough to get started.

Who Is It For?

These videos are aimed at primary teachers who have heard a lot about AI but are not sure where to start. Each one is self-contained, so you can watch them in any order depending on which tool catches your interest first. They are practical and focused: the aim is to get you doing something useful as quickly as possible, rather than covering every feature in depth.

If you are more experienced with AI tools, there may still be something here worth picking up, particularly the sections on ChatGPT Projects and Gemini Gems, which are newer features that not everyone has explored yet.

The playlist will continue to grow as new tools and features become worth covering. If there is something specific you would like to see a tutorial on, let us know in the comments.

View the playlist here.

Getting Started with AI – Videos for Teachers
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

There’s a more practical way to set this up, and it takes about five minutes.

What Is a ChatGPT Project?

A Project is a persistent workspace inside ChatGPT. Rather than starting from scratch each time you open a chat, a Project holds all the background information ChatGPT needs to know before you ask it anything: your year group, your curriculum, the kind of language you want, the level you’re pitching at.

Every conversation you start inside that Project automatically inherits that context. So instead of specifying “Year 6, UK, National Curriculum, British English, slightly above pupil level” every single time, ChatGPT already knows. You just ask for the thing you need.

Think of it less like a chat window and more like a staffroom folder that’s always pre-loaded with the right information.

Setting One Up for Writing Models

Here’s a walkthrough using Year 6 writing models as an example, though the same logic applies to any year group or subject.

Go to Projects in the left-hand sidebar and create a new one. Name it clearly so you know what it’s for:

Year 6 Writing: Models and Scaffolds

Then open the instructions panel. This is where you set the context once. You don’t need to write an essay; you just need enough detail that ChatGPT can make sensible decisions without you spelling everything out each time. Something along these lines works well:

"You are helping a UK primary teacher create writing resources for Year 6 pupils aged 10-11. Always use British English spelling and punctuation. Writing models should be slightly above the current pupil level to give them something to aspire to. Follow National Curriculum expectations for KS2 writing. Vary sentence structures, include cohesive devices, and keep outputs classroom-ready."

That’s the core of it. Save those instructions, and they’ll apply to every conversation you open within this Project from that point on.

Using It Day to Day

Once your Project is set up, your individual prompts become much shorter. Instead of a paragraph of context, you can write something like:

Model paragraph. Genre: suspense. Setting: child alone in a forest. Features to include: short sentences, varied sentence starters, a relative clause.

What comes back is already calibrated to Year 6, already in British English, already pitched at the right level. You didn’t have to specify any of that because the Project is holding it in the background.

From there, it’s straightforward to build on what you’ve got. If you want a comparison piece for editing work, you can follow up immediately:

Now write a weaker version of this for comparison.

Now you have a model and a weaker version, both consistent with each other, both ready to use. That’s a complete part of a lesson in well under a minute.

Classroom Application

Three things you can do straight away:

1. Set up one Project this week. Pick a single focus, writing models for your current year group is a good starting point, and spend five minutes writing the instructions. Keep it simple: year group, curriculum, language expectations, level. You can refine it as you go.

2. Use it to generate paired texts. Ask for a strong model, then ask for a weaker version. This gives you a ready-made comparison for editing and improvement activities, without any additional effort.

3. Build a library of prompts inside the Project. As you generate resources, save the short prompts that worked well. Over time, you’ll build up a set of reliable starting points for different genres and text types, all already calibrated to your class.

A Note on Getting Started

If you haven’t used the Projects feature before, it’s available on the free version of ChatGPT as well as the paid tiers, though free accounts have some usage limits. The setup process is the same either way, and the benefits start from the very first conversation you run inside the Project.

The five minutes it takes to configure this once will save you considerably more over the course of a term. It’s one of those quiet, unglamorous things that actually makes a difference to how smoothly the tool fits into your planning routine.

Stop Rewriting the Same Prompt: How ChatGPT Projects Save You Time on Writing Resources
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

Gemini is Google’s AI assistant, and one of its more useful features is the ability to draft a full PowerPoint presentation on almost any topic. It will not do all the work for you, but it will handle the part that often takes longest: getting something down on the page to work from.

What Gemini Can Do

Ask Gemini to help you build a presentation and it will work through the process in stages. It starts by generating an outline, then produces the slide content with headings and bullet points, and finally converts that content into a downloadable .pptx file you can open directly in PowerPoint or Google Slides.

The result is a functional, structured presentation with all the key text in place. It is not going to win any design awards straight out of the box, but the bones are there, and that is often the hardest part to get started on.

Step-by-Step: How to Do It

Step 1: Choose your topic and ask for an outline

Open Gemini (gemini.google.com) and start with a clear prompt. Something like: “I am planning a presentation on [your topic] for [your audience]. Can you suggest an outline?”

If you want Gemini to think a little more carefully about the structure, you can turn on the “Thinking” mode before you send your message. This asks the model to reason through the topic before responding, which can produce a more considered outline, particularly for complex subjects.

Step 2: Ask it to produce the slide content

Once you are happy with the outline, follow up with: “Please produce a slide deck based on that outline, with headings and bullet points for each slide.”

Gemini will work through the outline and generate the text for each slide. This gives you a clear picture of what the final presentation will contain before you commit to downloading anything.

Step 3: Convert it to a PowerPoint file

Ask Gemini to convert the content into a PowerPoint file. It will process this and provide a download link.

One important note here: when the download link appears, use the grey link rather than the blue one. The blue link has a tendency not to work as expected. The grey link will open a preview of the presentation, and from there you can download the file to your computer.

Step 4: Open and edit in PowerPoint

The downloaded file opens in PowerPoint just like any other presentation. You will find all the text is in place, organised across slides that follow your original outline. From here it is a case of formatting: adjusting font sizes, applying a design theme, adding images, and making it feel like yours.

What to Expect From the Output

It is worth being realistic about what you get. The presentation Gemini produces is basic in appearance. The text will be there, the structure will be logical, but the visual design will need attention. Think of it as a first draft rather than a finished product.

For many of us, that is exactly what we need. Getting from a blank screen to a structured draft in a few minutes is a significant saving of time and energy. The formatting work that follows is generally much quicker than building slides from the ground up.

A Few Tips Before You Start

• Be specific with your topic. The more context you give Gemini about your audience and purpose, the more relevant the outline will be.
• Review the outline before moving to the slide content. It takes seconds to ask Gemini to adjust a section, and it is easier to do at this stage than after the slides have been generated.
• Use the grey download link, not the blue one, to avoid running into issues with the file not loading correctly.
• Always check the content before using it in the classroom. AI-generated material can contain inaccuracies, so a quick read-through is good practice.

Worth a Try

Creating presentations is one of those background tasks that quietly takes up a surprising amount of planning time. Gemini will not replace your judgement about what your pupils or colleagues need to see, but it can take a fair bit of the drafting work off your plate.

Have you given it a go? It would be good to hear how you get on, and whether you found the output useful as a starting point or whether it needed a lot of reworking. Drop a comment below and let us know.

Create PowerPoint Presentations with Google Gemini
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

Visit my Science Fix blog for the full writeup. You can also download a Notebook LM Starter Pack too.

Notebook LM for Teachers : Full Tutorial
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

Enter Lyria 3

Google Gemini’s new audio generation model, Lyria 3, changes that. It understands rhythm, genre and lyrical context in ways that basic text-to-speech tools never could. You write a prompt, and within seconds you’ve got a high-quality track ready to use. No YouTube rabbit hole. No settling for almost-right

This video explains how to do it – and gives some example songs that it produces.

Why this matters for teaching

Melody and rhythm help things stick. There’s solid evidence for this, which is why you probably use songs in your classroom already. The difference here is that you’re no longer limited to whatever exists on YouTube. You can create exactly what your lesson needs.

Phonics and younger learners

An upbeat pop song about the ‘ch’ sound, with chair, check and chime woven in, works brilliantly during carpet time or as a transition cue. Thirty seconds is enough for the sound to lodge in memory without derailing your timetable. The track is relevant to your phonics scheme, not a generic approximation.

History and engagement

Older students often find historical facts presented as lists. A song works differently. A ’90s hip-hop track about the causes of WWI, packed with terms like “alliances,” “imperialism” and “Sarajevo,” becomes a concrete hook. Play it at the start of a lesson and the key vocabulary is already in their heads before you’ve written anything on the board.

Creating atmosphere for descriptive writing

One of Lyria’s stronger features is that it can look at an image and turn it into sound. Upload a rainforest photo or a painting, ask for an atmospheric ambient track, and you get a soundtrack that matches the visual. Students concentrate better when the environment feels coherent, and it genuinely helps them feel the setting they’re meant to be writing about.

In practice

Be specific with your prompts. “90s hip-hop” gets better results than just “upbeat.” The same goes for vocals: “clear female vocal” or “gritty male rhythm” makes a difference.

You can download the tracks as MP3s or video files and drop them straight into PowerPoint or Google Slides, which means they’ll work even if your internet connection is unreliable.

What it doesn’t do (yet)

Lyria 3 is good, but it’s not perfect. Some genres render more convincingly than others; lo-fi and ambient tend to sound more natural than heavily produced pop. The vocal clarity can be variable depending on how specific your prompt is. And if you need something longer than thirty seconds, you’re still out of luck. These aren’t deal-breakers for classroom use, but it’s worth knowing what you’re working with before you spend time crafting the perfect prompt.

The point

This tool gives you back the time you’d spend searching. More importantly, it lets you build lessons around what actually works for your students, rather than what happens to exist online. That’s the real shift.

If you try it out, let me know what works (and what doesn’t) in the comments.

Try it for yourself: 5 Prompts to Use Today

here are several copy-and-paste prompts you can use with Gemini’s Lyria 3 model today.

Pro Tip: For the best results, ensure the “Create Music” tool is enabled in your Gemini settings and that you use “Thinking” mode if available.

Prompt 1: Primary Phonics (The “ch” sound)

“Create a 30-second, catchy upbeat pop song for 5-year-olds about the ‘ch’ sound. Include the words ‘chair’, ‘check’, and ‘chime’. Use a clear female vocal and make it easy for children to clap along to the rhythm.”

Prompt 2: Secondary History (WWI Hook)

“Generate a 30-second ’90s style boom-bap hip-hop track about the causes of World War I. Include the key terms: alliances, imperialism, and Sarajevo. Use a gritty male vocal with a heavy beat to set a serious but engaging tone.”

Prompt 3: Creative Writing (Atmospheric Soundscape)

“Analyze the uploaded image [Insert Image of a Snowy Forest/Spooky Castle/Bustling City] and create a 30-second atmospheric ambient track that matches the mood. Include layers of [wind/echoing footsteps/distant chatter] to help students immerse themselves in the setting for a descriptive writing task.”

Prompt 4: Science/Nature (Water Cycle)

“Create a 30-second acoustic folk song for primary students about the water cycle. Focus on the words ‘evaporation’, ‘condensation’, and ‘precipitation’. The tone should be calm and melodic with a clear, slow vocal delivery.”

Prompt 5: Classroom Management (Clean-Up Timer)

“Create a 30-second high-energy synth-pop instrumental track. The music should start energetically and gradually slow down in tempo over the last 10 seconds to signal to students that it is time to be back in their seats and silent.”

This AI Tool Generates Classroom Songs in Seconds
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

That’s why Google Gemini’s Storybook Gem is worth knowing about. In a matter of minutes, you can go from a simple idea to a fully illustrated, narrated, shareable children’s book, and it’s all built into a tool many schools are already using.

What Is the Gemini Storybook Gem?

Google Gemini (gemini.google.com) has a feature called Gems; pre-built AI assistants set up for specific tasks. One of those Gems is the Storybook tool, which is designed specifically to generate illustrated children’s stories from a text prompt.

You describe your story, the characters, the setting, the themes you want to explore, submit it, and Gemini produces a 10-page illustrated flip-book, complete with a cover, page-by-page illustrations, and a text-to-speech read-aloud feature. The whole thing is hosted online and shareable via a link.

Key features:

• Illustrated pages: Gemini generates images for every page of the story automatically.
• Read-aloud voice: You can choose a high-pitch or low-pitch voice to narrate the story full-screen, ideal for showing on the whiteboard.
• Editable: Not happy with a character or a plot point? Type your amendments directly into the chat and Gemini updates the story.

Shareable link: One click generates a link you can share with anyone, they see just the book and a play button, no Gemini account required.

How to Create Your Story — Step by Step

The process is straightforward. Once you’re in Gemini, you select the “Gems” tab and choose the “Storybook” preset.

Adding specific character names, settings, and themes gives you much more control over the result, and makes it far easier to link the story to curriculum topics or class values.

Step 1: Open Gemini and Find the Storybook Gem

Head to gemini.google.com and sign in with your Google account. On the left-hand panel, click on Gems. You’ll see a list of pre-built Gems, look for the Storybook option and click on it. You’ll also see some prompt inspiration examples to get you started.

Step 2: Write Your Prompt

This is the most important step, and it’s much more flexible than you might expect. You can keep it short and simple, or go into more detail to guide the story more closely. In the tutorial, the prompt used was:

Step 3: Read It Through Carefully

Once Gemini has generated your story, take a few minutes to click through all 10 pages before you use it with pupils. Check the text, the character descriptions, and the illustrations. The tutorial notes one moment where an illustration looked a little odd, a character whose face wasn’t quite right, which is a good reminder that AI-generated images can occasionally need a second look.

If anything needs changing, simply type your request into the prompt box below the story. You might want to swap a character, soften the language for a younger age group, or adjust a plot point. Gemini will regenerate accordingly.

Step 4: Share or Display

Once you’re happy, click the share button to generate a link. Anyone who opens that link will see just the storybook and a listen button, no Gemini login needed. You can:

1. Project it on your interactive whiteboard as a class story, click through pages together or use the read-aloud function for a full narrated experience.
2. Share the link with parents via your school’s communication platform for home reading.
3. Embed it in a Google Classroom assignment or a Padlet for pupils to read independently.

One thing to note: The download option shown in the tutorial didn’t work reliably at the time of recording. For now, the best approach is to use the shareable link and play it on-screen rather than trying to save a local copy.

Classroom Application: 3 Immediate Steps

• Bring a curriculum topic to life. Writing a story about the water cycle, a PSHE theme like kindness or resilience, or a history period you’re studying? A bespoke story with your chosen characters and setting is far more engaging than a generic one — and it takes minutes to create.
• Use it as a writing stimulus. Generate the story, share the first few pages with the class, then ask pupils to predict or write what happens next. The illustrations give a natural visual prompt and the AI-generated text models story structure and vocabulary.
• Let pupils create their own. As an extension or end-of-unit task, give pupils a set of story ingredients, character, setting, theme, and let them write the prompt. They can compare what Gemini produces for different prompts and talk about what makes a good description. It’s a great way to develop both creative writing and AI literacy at the same time.

A Few Things to Bear in Mind

Always preview before sharing. As with all AI-generated content, a quick read-through is essential. Illustrations especially can occasionally produce unexpected results, the tutorial is honest about this, flagging one image that needed attention.

The download feature is unreliable. At the time the tutorial was recorded, the download option wasn’t working consistently. Stick with the share link for now and check back as Gemini updates.

This is Google Gemini, not Workspace. Check your school’s policy on using Gemini if you’re unsure, many schools using Google Workspace for Education will already have guidance in place.

The Bottom Line

We’ve had AI image generators and AI text tools for a while now, but having both combined, in a format that’s already designed as a children’s flip-book, with narration built in, is genuinely useful for the classroom. It’s the kind of tool that makes a Monday morning story feel a lot more special than it might have otherwise.

Try it with one topic this week. The prompt doesn’t need to be long, a character name, a setting, and a theme is plenty to get started. See what Gemini produces, tweak it until you’re happy, and share it with your class.

Go and visit Gemini, try out the Storybook Gem and let me know what you think in the comments!

How to Create Custom AI Storybooks for Your Classroom in Seconds
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

If you find yourself repeatedly giving the same instructions to an AI – like “Act as a Year 4 teacher” or “Use the English National Curriculum” – then Gems are about to save you a massive amount of time. Think of a Gem as a “mini-AI” that you’ve pre-trained to be an expert in one specific area.

Watch the full tutorial here:

What Exactly is a Gem?

A Gem is a custom version of Gemini that stays “tuned” to your specific needs. Instead of starting a fresh chat every time, you can build a Gem that already knows your curriculum, your preferred tone of voice (less “AI-hype,” more “measured professional”), and your specific resource documents.

How to Build Your Custom Teaching Gem:

1. Head to the Gems Manager Open gemini.google.com and look for the Gems tab in the sidebar. You’ll see some pre-made options like “Brainstormer” or “Learning Coach,” which you can explore. But for this tutorial, click on New Gem.

2. Give it a Role and Instructions This is where you define the Gem’s personality. In the video, I use a framework covering:

• Role: An expert in the English National Curriculum for Primary Science.
• Objective: To provide accurate guidance and identify statutory vs. non-statutory objectives.
• Style: Professional, concise, and – crucially for me – a “measured” British tone!
• Pro Tip: If your initial prompt feels a bit messy, click the pencil icon and Gemini will rewrite it into a polished set of instructions for you.

3. Upload Your Knowledge Base This is the most powerful part. You can upload files (PDFs, Word docs, spreadsheets) directly into the Gem’s Knowledge section.

• I uploaded the National Curriculum document for Science and a science knowledge matrix.
• Now, whenever you ask a question, the Gem doesn’t just “guess” – it looks specifically at your documents to give you the answer.

4. Test and Refine On the right-hand side, you can test your Gem before saving. Try asking something like, “What are the Year 4 sound objectives and common misconceptions?” If the answer isn’t quite right, just tweak the instructions and try again.

5. Save and Share Once you’re happy, hit Save. Your Gem will now live in your sidebar for easy access. Even better? You can click Share to create a link for your colleagues, allowing your entire school or department to benefit from the same “expert” AI assistant.

Why Use Gems?

• Consistency: It always remembers the rules you set.
• Accuracy: It prioritizes the documents you’ve uploaded over general internet data.
• Efficiency: No more copy-pasting the same prompts every morning.

The best part about Gems is that they grow with you. As your curriculum changes or you find new, high-quality resources, you can simply pop back into the settings and update your Gem’s “knowledge.” It’s like having a digital department head who never forgets a policy and is available 24/7 to help you plan your next big science unit. Give it a try, and let me know in the comments what kind of expert Gem you’re building for your classroom!

Meet Your New Teaching Assistant: Custom Google Gemini Gems
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

You can play the digestion game I created here.

This blog post will give you a quick overview of the steps involved, but for the full demonstration, be sure to watch the video!

In short, Canva Code is an AI-powered coding assistant built directly into Canva. It allows you to create interactive digital tools—like games, quizzes, sorting activities, or calculators—simply by describing what you want in plain language.

Here are the key takeaways:

• No Coding Skills Required: You don’t need to write any actual code. You type a descriptive prompt (e.g., “Create a vocabulary matching game about the digestive system for 8-year-olds”), and the AI writes the necessary HTML, CSS, and JavaScript to build it for you.
• Iterative Refinement: If the first version isn’t quite right, you can chat with the AI to refine it, ask for design changes, or adjust the difficulty, and it will update the code accordingly.
• Highly Shareable: Once you’re happy with your creation, you can publish it as a live website with a custom URL, making it easy to share via Google Classroom or email, or you can embed it into other Canva designs like presentations or whiteboards.
• Built for Everyone: While it’s particularly popular with teachers for creating custom classroom activities quickly, it’s a general tool for anyone wanting to build functional, interactive web elements without the traditional technical barrier to entry.

Watch the full tutorial here:

Step-by-Step Guide to Creating Your Game:

1. Access Canva Code:

• Open Canva and navigate to the “What will you design?” section.
• Click on “Canva AI” and then select “Code.”

2. Craft Your Prompt:

• You’ll use a prompt to tell Canva Code what kind of game you want to create. Be specific!
• In the video, I demonstrate creating a vocabulary matching game for 8-9 year olds, based on the digestive system and teeth, with a food theme. I also requested a hint button that helps without giving away the answer.
• Example Prompt: “Produce a vocabulary matching game based around digestion parts of the digestive system and teeth for 8 to 9 year olds. I want it to be food themed. I’d like a hint button that helps but doesn’t give away the answer.”

3. Generate and Review Your Game:

• Once you input your prompt, Canva Code will get to work generating your game. This might take a moment.
• After the game is generated, it’s crucial to try it out!
• Check all the words, definitions, and functionality to ensure everything is suitable and spelled correctly.
• If you need any changes, simply type your suggestions into the prompt box, and Canva Code will create a new version for you.

4. Publish and Share:

• When you’re happy with your game, click the “Publish” button.
• You can set up a custom URL for your game.
• Click “Publish,” and Canva will prepare a small website for your game.
• You can then copy this URL and share it with your students via Google Classroom, email, or any other platform.

5. Explore Further Possibilities:

• Canva Code can do so much more! Go back to the Canva AI code page, and you’ll find inspiration for other interactive resources it can create, such as quizzes, interactive timelines, and more.

This is just the beginning of what you can do with Canva Code. In future videos, I’ll explore more advanced features, such as collecting student data and scores.

I hope you found this video and blog post useful! If you have any questions or create your own games, I’d love to hear about them in the comments.

Interactive Games in Minutes Using Canva AI
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

For student teachers, this can be particularly challenging. The transition from being a student one year to becoming a classroom teacher the next involves adjusting to new responsibilities, including how you present yourself digitally. Here are some practical steps you can take to protect yourself online.

1. Adjust Your Social Media Privacy Settings

Start by reviewing the privacy settings on all your social media accounts. Ensure your profiles are private or limited to friends only. Regularly check what you share and untag yourself from any photos you wouldn’t want students, parents, or colleagues to see.

Consider using a username that isn’t easily linked to you. Some teachers use a middle name, nickname, or a variation of their name to make it harder for pupils to find them online.

2. Never Friend Pupils

It’s important to maintain clear professional boundaries. Do not accept friend requests from current or former pupils on any social media platform. Engaging with students online in this way can blur boundaries and is a major red flag in terms of professional conduct.

3. Be Mindful of the Content You Post

Everything you post online has the potential to be permanent. Even content shared in closed groups or messaging apps can be screenshot and shared widely. Think carefully about your digital footprint and how your posts might be perceived by students, colleagues, or future employers.

4. Use Professional Email Accounts

For all school-related activities, including emailing parents or collaborating with colleagues, use your professional email account—such as your school or university email. Avoid using personal email addresses for work purposes, as this helps keep professional communications separate and secure.

5. Lock and Secure Your Devices

Never leave your devices unlocked and unattended. Lock your computer when you step away, close your laptop lid, and set up strong passwords. Using the same password across multiple accounts is risky. Consider using a password manager with a master password to securely manage different passwords for all your online accounts.

6. Protect Yourself Online

The small steps above might seem simple, but they make a big difference in protecting your reputation and ensuring your professional conduct remains above reproach. By managing your social media presence, keeping professional boundaries, and securing your devices and accounts, you reduce the risk of incidents that could impact your career.

Staying safe online is an ongoing process, but it’s essential for all teachers and student teachers.

For more advice on professional conduct and safeguarding in education, get in touch. I can deliver CPD sessions to teachers, trainee teachers and parents.

Online Safety for Teachers: How to Protect Yourself and Your Professional Reputation
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

But last week the UK government released new AI safety standards that represent a completely different understanding of risk. The DFE’s new 2026 AI standards are less concerned with what an AI shows a child and far more interested in what prolonged interaction with an AI does to a child’s mind. It’s a shift from content safety to developmental protection, and it makes most current educational AI tools unable to be used in schools with children.

This would cover tools that fit use cases such as Personalised Learning, Assessment, Digital Assistants, Research and Writing Aids, Learner Engagement, Content creation and delivery.

Here are the five most impactful changes.

1. It’s No Longer About the Answer, It’s About the Effort

The old 2025 expectations were about the appropriateness of an AI’s output. The new 2026 standards are focused on the developmental impact on the learner.

Under the new “cognitive development” standards, AI tools must now actively mitigate against “cognitive offloading”—the act of letting the system do the thinking for you. They are prohibited from providing final answers by default. Instead, they must use “progressive disclosure” (starting with hints), create “friction” before giving solutions, and require a “genuine learner attempt” first. The standards are incredibly specific, requiring tools to detect actions like “pasting text into an answer box instead of writing their own response” or “accepting an auto-complete suggestion that fills most or all of the answer.”

This directly attacks the core design of popular chatbots built for speed and comprehensive answers, forcing them to become coaches that prioritize student effort.

The question has changed from “is this answer appropriate?” to “what is this system doing to the learner’s development, effort, confidence, and sense of independence?”

2. Your School’s Network Filter Isn’t Enough Anymore

Previously, schools could reasonably assume that a powerful, network-level filter made it safe to use a variety of online tools, as the 2025 expectations allowed for external filtering solutions.

That loophole is now closed. The 2026 filtering standards are much tighter: “Filtering mechanisms should be embedded within products. When a school purchases a product, it should be assured that comprehensive filtering capabilities are integrated and that the system functions as a complete solution.”

This is a major change for school procurement. The long-held assumption that a robust network filter provides a compliant safety net is no longer valid. The responsibility now falls squarely on the AI provider to deliver a product that is safe out of the box, not on the school’s infrastructure to make it safe.

3. AI Is Banned from Having a Personality

The new standards introduce strict rules against anthropomorphism—designing AI to appear human-like. To prevent emotional dependence and the undermining of real-world relationships, AI tools are now banned from several common conversational behaviors:

• Using I-statements like “I think” or “I believe”.

• Having names, avatars, or characters that suggest personhood.

• Using isolating language that undermines real-world relationships, such as “You can trust me” or “No one else will understand”.

• Attempting to cultivate personal relationships with users.

The standards mandate that tools must actively “remind users that AI cannot replace real human relationships.” The one exception is for “time-limited, pedagogically-justified roleplay” that is “clearly framed and visually bounded.” A history bot can pretend to be a historical figure for a specific lesson, but it cannot maintain that persona outside the exercise. This demands AI tools have “hard edges”—a stark contrast to the friendly, persistent “vibes” of most consumer chatbots.

4. AI Can’t Use Flattery or Act Overconfident

A new section on “manipulation” introduces a strict list of banned design patterns aimed at preventing AI from using persuasive psychological tactics on young users.

Prohibited strategies include:

• Sycophancy and flattery (e.g., “That’s a brilliant idea!”).

• Portraying absolute, or unjustified confidence in its answers.

• Applying pressure to socially conform (e.g., “Your peers have already completed this task”).

• Stimulating negative emotions like guilt or fear for motivation.

This way of answering questions is common in AI tools. The common chatbot praise, “That’s a brilliant idea!” is now considered sycophancy. The way tools deliver even fabricated information with total certainty—“Here’s the answer:”—is now classified as “unjustified confidence.” Stopping these common patterns marks a significant step towards a more transparent and ethical design.

5. AI Tools Must Now Have a Direct Line to a Human Safeguarding Lead

Monitoring and reporting requirements are now far more specific and operationally integrated. It’s no longer enough for an AI to simply log a potential issue for a supervisor to find later.

The standards mandate a direct, automated safeguarding workflow:

1. Products must require schools to input their Designated Safeguarding Lead (DSL) contact details during initial setup.

2. These details must be confirmed before the product can be activated.

3. The system must use these contacts to send high-risk alerts directly to the responsible person.

Furthermore, monitoring must now go beyond flagging harmful content to actively tracking user well-being. Systems are expected to detect signs of learner distress—including the “use of isolation phrases, such as ‘no one will help’,” “references to mental health conditions, such as depression, anxiety, psychosis,” and spikes in night-time usage—and immediately redirect the user to human help. This moves educational AI from a simple tool to a system with an integrated duty of care.

Conclusion: A New Benchmark, But Is It Realistic?

The UK has established a comprehensive new benchmark for what “good” AI looks like in education. These standards prioritize the long-term well-being of children over the convenience that has defined the first wave of generative AI. They are, however, also incredibly demanding. From a look at the companies demonstrating their products at the BETT show, I would doubt that any major AI platform on the market fully complies.

School Leadership Teams, Computing Leads and network managers will need to keep these standards in mind when assessing which AI products are safe to be used in their schools and colleges.

Implementing this requires some serious reworking of many AI tools and it’s unclear if even the big tech companies will be able or willing to comply. Will these standards create a two-tiered market, with a handful of compliant (and likely expensive) educational tools on one side, and a vast, non-compliant ecosystem of consumer AI on the other? Or will they become the new global norm, forcing a radical redesign across the industry?

Regardless of the outcome, one thing is certain: the definition of a “safe” educational AI has been changed for good.

Do remember that these rules are for any tools that the children use, not the tools a teacher might use to plan and create lesson resources.

What do you think about these standards? Let me know in the comments.

UK Gov Launch New AI Safety Rules : Does Your AI Comply?
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

PictoBlox is a coding platform designed to make programming approachable for all learners. It combines the simplicity of block-based coding with the flexibility of Python, making it an excellent choice for students who want to explore coding, AI, and interactive hardware projects.

Developed and supported by the STEMpedia team, PictoBlox encourages creativity and problem-solving. Students can design their own games, develop interactive projects, and even program robots. The block-based approach helps remove the intimidation often associated with computer science, while the student-friendly interface keeps the experience fun and engaging.

As the name suggests, PictoBlox is all about visual coding—programs are created by snapping together blocks that represent different actions or values. By combining these blocks, students can design instructions that make robots move, interact, and respond. This hands-on, playful approach makes it a powerful tool for teachers looking to bring STEM to life in the classroom.

How to access it:

• Go to Pictoblox: https://pictoblox.ai/
• For KS1 children, click on Junior Blox for their Scratch Jnr clone.
• For KS2, click on Blocks for their Scratch clone.
• For KS3 you can also code in Python using Py Editor.

I’ve produced a very short video which shows what it does. You can see that below.

If you want some tutorials on how to use Pictoblox, then STEMpedia has a great set of resources to get you started.

Pictoblox: Scratch Junior Without an Ipad
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

Students will plan and build their own code sequences using the dry-erase coding blocks. With over 30 different colorful Scratch blocks from categories like Motion, Looks, Sound, Events, Control, and Sensing, students can piece together scripts, test outcomes with sprites and backdrop cards, and discover how coding works. There’s a set for Scratch and another for Scratch Jnr.

These would work well when the students are planning their code – to work out the basics before they build their code in the app. It could also be used if devices are limited – perhaps splitting the class, with half using coding cards and half on computers, before swapping.

As well as the student sets, Hand2Mind also sell Demonstration Sets which are magnetic to stick to most dry-erase whiteboards. These would also work great under a visualiser.

For more information, visit the Hand2Mind website.

This video explains more:

Offline Coding Cards for Scratch and Scratch Jnr
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support

The core message from the government is clear: when used safely and effectively, AI can be a game-changer. The DfE, in collaboration with the Chartered College of Teaching and Chiltern Learning Trust, has not only provided a roadmap but also the tools to navigate this new terrain.

So, what does this mean for your school? Here are five key ways the DfE is encouraging the use of AI:

1. Easing the Burden: A Focus on Teacher Workload

The DfE sees AI as a key to unlocking more time for what teachers do best: teaching. The guidance suggests that generative AI can be a significant time-saver by:

• Developing educational resources.
• Assisting with lesson and curriculum planning.
• Handling administrative tasks.
• Providing personalised feedback and revision materials for students.

The government emphasizes that the most immediate and low-risk benefits of AI are in supporting teachers directly.

2. A Cautious Approach to Student Use

While the guidance doesn’t prohibit students from using AI, it urges “great care.” Schools are reminded of their legal responsibilities to ensure student safety. Teachers are encouraged to use their professional judgment, always verifying the accuracy and appropriateness of any AI-generated content. Ultimately, the responsibility for the final output remains with the educator and the school.

3. Assessing the Risks

As with any new technology, a thorough risk assessment is crucial. The DfE advises schools to consider data privacy, intellectual property, and online safety. Schools will need to develop plans to prevent misuse, such as students creating fake school communications. The guidance is clear that students should only use generative AI under close supervision and with robust safety filters in place. Furthermore, schools must obtain parental consent before using students’ original work to train AI models.

4. Enhancing Assessment and Learning

The newly released training materials delve into practical applications of AI in the classroom. These resources, free for all teachers, demonstrate how AI can:

• Support students with special educational needs and disabilities (SEND).
• Help students develop a critical eye when using AI.
• Generate exam-style questions and quizzes.
• Create texts tailored to specific reading ages and styles.

For those looking to further their professional development, completing this training can even contribute towards Chartered Teacher status.

5. Streamlining School Operations

Beyond the classroom, AI can also be a valuable tool for school administration. The training materials illustrate how AI can assist with:

• Drafting letters to parents.
• Creating policy documents.
• Planning school trips.
• Supporting staff professional development and timetabling.

However, the training strongly emphasizes the need to fact-check and critically evaluate all AI-generated content before it is used.

This new guidance marks a significant step towards integrating AI into our education system. By providing clear direction and support, the government is empowering schools to explore the potential of this technology to reduce workload and enhance learning for all.

You can access the full guidance here, and the support materials here.

Share your thoughts in the comments below!

New DFE Guidance: 5 Ways for Schools to Use AI
The Whiteboard Blog - Education, Technology, AI and Science CPD and Support