How do we keep AI learning safe for young people?

AI is changing the world young people are growing up in. Helping them to understand how it works, question its outputs and use it responsibly is an important part of preparing them for the future. But how can schools introduce AI safely in the classroom?

At Apps for Good, we believe every young person should have the skills, confidence and opportunities to shape their future. Our free AI for Good 2.0 course, created by experienced educator and digital curriculum leader Jennifer Talbot, helps students explore AI safely and responsibly through practical, project-based learning. Students develop their AI literacy while creating solutions to real-world problems.

Safety is central to this learning. Students aren't given unrestricted access to public AI chatbots. Instead, they explore AI through structured, age-appropriate activities with clear learning objectives and teacher supervision.

Throughout the course, students investigate topics including bias, deepfakes, misinformation and data privacy. They learn that AI can produce inaccurate or misleading information, and that its outputs should always be questioned and checked.

Where students use generative AI directly, this takes place within Canva Education using school- or teacher-managed accounts. Canva states that student data is not used to train its AI models and provides safeguards within its education platform.

Other activities, such as the Prompt Battle, use a teacher-operated large language model (LLM), so students don't need their own accounts.

Students remain at the centre of their learning. They develop their own ideas, design solutions, write and test code, respond to feedback and evaluate their work. AI supports selected stages of the process but doesn't replace students' own thinking or creativity.

Our approach to safe AI learning is built around a few key principles:

• Protecting personal information: Students must never enter personal or sensitive information into AI prompts, including names, addresses, photographs of identifiable people, health information or details about their personal lives.
• Teacher supervision: Teachers oversee activities, review prompts and outputs, and decide what is appropriate to share with the class.
• Critical thinking: Students learn to recognise AI's limitations, question its responses and check for inaccuracies, assumptions and bias.
• Responsible use: Students are not encouraged to use AI for personal or emotionally sensitive conversations. They should also avoid entering confidential or substantial copyrighted material into AI systems.
• Safeguarding: If a tool generates unsuitable or concerning content, teachers should stop displaying it and follow the school's usual safeguarding and reporting procedures.

The Department for Education's January 2026 Generative AI product safety standards highlight important principles for using generative AI safely in education.

AI for Good 2.0 reflects relevant principles through its clear educational objectives, teacher supervision, age-appropriate and task-focused activities, data minimisation and emphasis on independent thinking. AI is not used to profile or grade students, or to make decisions about their educational outcomes.

The Prompt Battle is a short, teacher-led activity that shows students how the wording of a prompt can influence an AI-generated response.

Students suggest prompts for a harmless, curriculum-related task. The teacher reviews suitable suggestions, enters selected prompts into a school-approved LLM and shares the checked outputs with the class. Together, students compare the results, identify inaccuracies or bias and suggest improvements.

Students don't need their own AI accounts, and no personal or identifying information is included in prompts. If the school hasn't approved a suitable tool, teachers can use pre-generated examples instead.

The activity helps students understand that AI responses depend on the instructions they receive, and that a confident-sounding answer isn't necessarily correct.

AI literacy means more than knowing how to use the technology. It means understanding its limitations, questioning its outputs and making informed decisions about how it's used.

Through AI for Good 2.0, young people develop these skills by exploring real-world challenges, investigating the ethical questions surrounding AI and creating their own app ideas.

Because preparing young people for the future means giving them not just the skills to use AI, but the confidence to question it, challenge it and decide how it should be used.