inFormal Intelligence

These are all AI systems. They learn from data, recognise patterns, and make predictions. Language models like ChatGPT or Gemini work on the same basic principle, just applied to text, conversation, and content creation.

Text is the most common barrier in youth work. Activity sheets, programme descriptions, application forms, briefing documents, social media posts, almost everything we produce relies on written language, and most of it is written at a level that excludes a significant chunk of the people it is meant for.

This workflow is the most universally useful one in the toolkit. It requires no special tools, no accounts, and no technical knowledge. All you need is an LLM and the text you want to adapt.

Give an LLM a piece of text and ask it to:

Each of these can be done in seconds. And they can be chained: ask for a plain language version first, then ask it to reformat that version for ADHD. Each step builds on the last.

Have your text ready. That is it. Any LLM works for this workflow.

One practical note: you do not always need to copy and paste. Most LLMs can accept an uploaded Word document or PDF and work with it directly. You can ask them to rewrite, reformat, or simplify the whole document and return it to you in a usable form. This is especially useful for longer materials like programme guides or information packs.

One caveat worth knowing: when you upload a document and ask an LLM to edit it, formatting often gets lost or simplified in the output, especially with larger files. Tables, columns, and visual layouts may not survive the process intact. For anything where the formatting really matters, it is sometimes cleaner to paste the key sections of text rather than uploading the whole file, and then reapply the formatting yourself at the end.

Each template below is ready to copy. Replace the text in brackets with your own content.

If you are working with a longer document, like a programme guide, an application form, or a project brief, you do not have to process it section by section. Upload the whole file directly to your LLM and give it a clear instruction. For example:

Not everyone can access information through text. For young people who are deaf or hard of hearing, for participants with dyslexia or low literacy, or simply for someone who learns better by listening, audio can be the difference between being included in your programme and being left out of it.

This workflow covers two complementary skills: turning speech into text, and turning text into speech. Together they open up a new range of formats for your materials without requiring specialist equipment or technical expertise.

Speech to text tools convert spoken audio into written text. In a youth work context this is useful in more ways than people often realise: transcribing a recorded session so participants can read it back, creating written versions of verbal instructions, making your meetings and workshops searchable and shareable, or helping participants who find writing difficult to contribute verbally and still produce a written output.

Text to speech tools take written text and convert it into spoken audio. For your participants this means you can offer an audio version of any written material: activity sheets, welcome messages, programme descriptions, briefing documents, reflection prompts. For people who struggle with reading, having the option to listen instead is not a small thing.

There are many TTS platforms available. ElevenLabs is currently the most sophisticated, with highly natural-sounding voices, multilingual support, and a lot of control over tone and delivery. Their free tier is limited in the number of characters you can generate per month, but it is worth trying for shorter outputs. Link: elevenlabs.io

For a more generous free option, Google has something that most people do not know about.

Google AI Studio gives you access to many of Google's top-tier AI functions for free, including a high-quality text to speech tool. Here is exactly how to use it.

These tools are most powerful when they become a normal part of how you prepare and deliver your programmes, not a one-off experiment. Here are some concrete ways to use them:

Good visuals make content more accessible. A well-designed poster communicates faster than a paragraph of text. An infographic breaks down a complex process so anyone can follow it. An illustrated explainer works across language barriers in a way that dense written content never will. AI image generation has made it possible to produce this kind of material quickly, without a designer, and without a budget.

Prompting for images is not the same as prompting for text. When you ask an LLM to rewrite a paragraph, you are working with language about language. When you ask it to generate an image, you need to build an entire scene in words: who is in it, what they are doing, where they are, what the light looks like, what angle you are seeing it from, what style it is rendered in, and what text should appear.

The more layers you describe, the closer the result will be to what you imagined. Vague prompts produce generic results. Specific prompts produce useful ones.

A few principles from Google's official prompting guide apply here: be specific and provide concrete details on subject, lighting, and composition. Use positive framing and describe what you want rather than what you do not want. Control the composition using photographic and cinematic language like "low angle shot", "aerial view", or "soft natural light". And iterate, refining images with follow-up prompts rather than starting over each time.

The prompts below are built using Google's official image prompting templates, adapted for youth work contexts. Each one follows the same principle: describe the scene in full, do not just list keywords.

Ever seen a visual you loved and wished you could create something similar? You do not need to figure out the prompt from scratch. Upload the image to ChatGPT and use this prompt:

ChatGPT will return a detailed text description of the image broken down into all the elements that matter: subject, composition, lighting, style, colours, mood, and any text present. You then take that description, swap in the things you actually need, and use it as your starting prompt.

This works in two ways. In the short term, it gets you much closer to the result you want without trial and error. But it is also one of the fastest ways to learn how to write image prompts, because you see exactly how an AI translates a visual into language. After doing this a few times with different types of images, patterns start to click and your own prompts get noticeably better.

Canva is a browser-based graphic design platform that lets anyone create professional-quality visual materials without any design experience. You work from templates, drag and drop elements, and export finished designs in whatever format you need. It is one of the most widely used tools in the nonprofit and youth work sector for a reason: it removes the barrier of needing a designer for every poster, report, or social media post your organisation produces.

The range of content types you can create in Canva is genuinely wide. Presentations and slide decks, social media posts and stories, event posters and flyers, programme reports and infographics, certificates, newsletters, video content, email banners, logos, activity cards, workshop materials, and much more. There are hundreds of thousands of templates across all of these categories, many of them already designed with clear, accessible layouts.

Canva's AI tools sit under the Magic Studio banner and are available within the editor. These are the ones most directly useful for youth work and inclusion-focused content creation:

In April 2026 Canva announced its most significant update to date: Canva AI 2.0. The shift moves the platform from a design tool with AI features to something much closer to a full creative collaborator. You describe what you need in plain language, Canva plans and executes the work across multiple formats at once, remembers your brand and style preferences, and can connect with tools like Google Drive, Gmail, and Notion to pull in context automatically.

Canva AI 2.0 launched as a research preview in April 2026 and is rolling out to users progressively. By the time you read this, it may already be available in your account. For youth workers producing regular visual content, the time savings are going to be significant.

These workflows go a step further than the previous ones. They combine multiple tools in sequence and produce more complex outputs. None of them require technical skills to get started, but they do reward experimentation and iteration.

This three-step workflow lets you go from a rough visual idea to a working, clickable web interface without writing a single line of code yourself. It is particularly useful for youth workers who want to build a simple landing page, a registration form, an information hub, or a basic interactive tool for their participants.

This workflow is excellent for simple, standalone interfaces: a landing page, an information screen, a registration form, a basic tool with a few inputs and outputs. It produces real, working HTML and CSS that behaves in a browser.

For more complex functionality, things like user logins, databases, stored data, or connections to external systems, Google AI Studio can technically generate and connect backend logic too. But this is where things get significantly more difficult. It is very possible to do, but the complexity rises fast, the risk of things breaking quietly or behaving insecurely goes up just as fast, and for someone without a software development background it becomes a long and frustrating journey. The tool will generate something that looks like it works, but making it actually work reliably and safely in a real environment requires understanding what is happening under the hood. If you want to build something at that level of complexity, involve a developer, or invest real time in learning the basics of how web applications are built before directing AI to build one for you.

Most people who use Google products every day have no idea that two of the most powerful and accessible AI platforms Google offers are sitting there for free, waiting to be used. Google AI Studio and Google Labs are where a lot of the tools mentioned throughout this toolkit actually live, and knowing they exist opens up a significant amount of additional capability for anyone willing to spend a few minutes exploring.

Google AI Studio is a free, browser-based workspace where you can work directly with Google's most advanced Gemini models, no setup, no installation, no technical background required. You sign in with any Google account and you are in.

It is the backstage of a lot of what this toolkit covers. The TTS workflow in Workflow 2, the image generation with Nano Banana, the prototype export from Google Stitch, all of these connect back to AI Studio. But it is also a powerful standalone environment for testing prompts, uploading documents, processing audio and video files, and building simple AI-powered tools.

The interface is always free. Where costs can apply is if you connect a Google Cloud billing account and start using the API at high volume, but for the kind of exploratory and practical use this toolkit is about, you are unlikely to hit those limits.

One thing worth knowing: on the free tier, Google may use your inputs to improve its models. If you are working with sensitive participant data or confidential organisational materials, use the paid tier or keep that content out of the platform entirely.

Google Labs is Google's public hub for experimental AI tools, the place where new ideas get tested with real users before they become mainstream products. Almost every major Google AI feature you use today started here. NotebookLM was a Labs experiment. Google Stitch was a Labs experiment. The Audio Overview in NotebookLM, Gemini Live, AI Overviews in Search, all of them spent time in Labs before most people knew they existed.

The entry point is labs.google. What you find there changes regularly as tools graduate to full products, get retired, or get replaced by something newer. It is worth visiting occasionally just to see what is available. Below are the tools currently in Labs or recently graduated from it that are most relevant for youth work and inclusive content creation.

Sometimes the most memorable way to raise awareness or teach something is through a song. Suno is an AI music generation tool that creates original songs complete with vocals, instrumentation, and lyrics from a simple text prompt. You describe the style, mood, topic, and what you want the song to say, and it produces something you can actually listen to, share, and use.

For youth workers this opens up a genuinely fun and effective avenue. A short catchy song about digital rights, online safety, inclusion, or any topic you are working on lands differently than a handout or a slide. It works well as an icebreaker, as a discussion starter, as a memorable closing to a session, or as shareable content for social media campaigns that cuts through in a way that text rarely does.

Most AI tools draw on everything they were ever trained on. Ask ChatGPT a question and it pulls from a vast, general knowledge base that you have no visibility into and no control over. That is useful for many things, but it creates a specific problem: you can never be fully sure where the answer came from, and the AI can confidently give you something wrong.

NotebookLM works differently. It only ever uses the sources you upload. Nothing more. You give it your documents, your research, your materials, and it works entirely within that boundary. Every answer it gives you includes a direct citation you can click to verify exactly where in your sources that information came from. This is what Google calls source-grounded AI, and it is the reason NotebookLM has become one of the most genuinely trusted AI tools for learning and teaching.

For youth workers this matters in two practical ways. When you are learning something, you can upload the materials you are working with and have a conversation with them, ask questions, request summaries, get explanations at different levels of complexity, without the AI wandering off into things you did not ask about. And when you are teaching or facilitating, you can build a notebook around your programme materials and use it to create multiple formats of the same content for different learning needs, all grounded in what you actually wrote.

NotebookLM organises everything into notebooks. Each notebook is a self-contained workspace with its own set of sources and its own chat history. You can have up to 100 notebooks on the free plan.

Once you have your sources uploaded, the Studio panel (accessible from the top of the notebook) lets you transform your content into a range of different output formats. This is where NotebookLM becomes genuinely powerful for inclusion work, because the same source material can become audio, visual, interactive, or structured depending on what your participants need.

This is where it gets interesting. Each skill above is useful on its own. But put two or three of them together and the outputs become something you could not have produced before without a team, a budget, and weeks of work.