Vocational Innovation – Teaching
Interactive learning material · M3-T1-MLC1
One shared goal – different support, steps and pace
Learning objective
Learners in vocational classes start with different knowledge and learn at different speeds. This material shows how AI can help teachers design personalised learning paths.
By the end of the learning content, teachers can use AI to describe and evaluate student profiles on a specific topic, as well as modify and approve the AI-suggested route so that their own pedagogical aspects prevail.
Basics · A shared goal and different starting points
Picture a group of first-year car mechanic apprentices. One apprentice has spent two years helping in a family garage, another has just finished general secondary school, a third is an adult career changer, and a fourth is still learning the language of instruction. They must all reach the same level of professional competence. Yet the same sequence of lessons will bore some and leave others behind.
A personalised learning pathway is a planned sequence of learning activities, learning materials, practice tasks, and checkpoints leading to a defined learning outcome, taking account of the learner’s starting point and progress. The approach has longstanding educational foundations: in 1984, Benjamin Bloom reported studies in which individual tutoring, supported by regular assessment, feedback, and targeted practice, proved substantially more effective than conventional classroom instruction. His findings underscore how important ongoing support tailored to the learner’s progress can be.
The problem has always been feasibility – no teacher can plan twenty individual routes by hand.
With generative AI, teachers can quickly create task sequences and explanations tailored to learners’ differing levels of prior knowledge and pace of learning. This can significantly reduce preparation time, but the generated materials are suggestions: teachers still need to set aside time to review their accuracy and educational suitability.
Basics · Outcome, prior knowledge and activities
Before asking AI for anything, it helps to know what a good path contains. Whether it is drawn on paper, built in Moodle or set up in a learning platform, a learning path usually has these elements:
Taken from the qualification or curriculum – the same for every learner.
what the learner already knows and can do, identified through a short diagnostic task, observation of practical work or a conversation.
A sequence of steps that builds on learners’ prior knowledge: for example, when learning to calculate a discounted price: reviewing the basics of percentages → working through an example together → practising with guiding questions → solving problems independently → checking answers and correcting mistakes.
Basics · More building blocks of a well-designed path
Resources in several formats – text, video, worked example, simulations, demonstrations – in line with the principle of multimodal learning.
Practice tasks with feedback, ideally in realistic workplace contexts and with increasing difficulty.
Checkpoints and decision rules: If successful, the learner can move on or receive a more challenging task; otherwise, they receive additional support.
A flexible pace and some learner choice, so that learners gradually take responsibility for their own learning.
Design with AI · Sequencing activities
AI can break a learning outcome into sub-skills, identify prerequisites and propose a logical order of activities. For an electrician lesson on residential circuits, for example, it can suggest moving from safety rules and circuit diagrams through measuring with a multimeter to wiring and testing a circuit. The teacher checks whether the order matches the curriculum, the workshop schedule and the available equipment.
Design with AI · Building on existing resources
AI can suggest learning materials at different levels and in different formats: brief explanations for beginners, step-by-step worked examples for learners who need help following the solution process, technical data sheets for advanced learners, and illustrated glossaries for students who need reading support or are learning the language. The accuracy and accessibility of sources recommended by AI need to be checked. It is therefore useful to provide materials you already know and trust – your own resources, content available in Moodle, textbook chapters or manufacturers’ manuals – and ask AI to select the sections relevant to the learning objective and arrange them in a sequence that supports learning. Ask it to create new explanations or practice tasks where the existing materials need to be supplemented.
Beginners work through the core concepts needed for the goal using a clear, brief explanation.
Learners who need help following the solution process receive a step-by-step worked example and guiding questions.
Advanced learners can use technical data sheets; learners who need language support may benefit from an illustrated glossary.
Different formats and forms of support lead to the same learning outcome.
Design with AI · Checkpoints and route changes
AI is good at producing many variants of practice tasks: graded calculation tasks for a welder estimating material, menu-costing cases for a cook, troubleshooting scenarios for an IT technician, short care situations for a nursing assistant. It can add hints for the support route and more open, realistic problems for the extension route. Every learning pathway should include short tasks that prompt learners to recall previously learned material from memory at regular intervals.
The heart of personalisation is the decision at each checkpoint. AI can help formulate clear decision rules – what happens if a learner scores low on the entry check, finishes early or gets stuck at the same step twice – and suggest alternatives for learners who need a different pace or extra language support.
The route can change as the learner progresses: they may receive additional support, continue on the standard route or move on to a more challenging task. The teacher regularly reviews the choice together with the learner.
Try an example
Select a route to see the support or tasks it might offer. This example illustrates a fictional learning situation.
Support loop
An illustrated glossary or a subtitled video supports understanding. A teacher demonstration is followed by guided practice. The learner tries again and can move to a different route at the next checkpoint.
Standard route
The learner practises using the specified, checked resources, then completes an independent task and receives feedback. If they get stuck, they can receive support; if they progress confidently, they can choose an extension task.
Extension
The learner tackles a more complex workplace problem and explains their solution. The extension route also leads to the shared learning outcome and final task.
A shared goal and final task – the route can change at the next checkpoint.
Car mechanic example · Starting point and profiles
A teacher plans a three-week unit on testing the battery and charging system for a group of 16 car mechanic apprentices. A short entry check (five questions and one observed measurement) shows three typical profiles. The teacher describes them functionally – no names, no grade lists, no personal data – and asks a school-approved AI tool for a draft.
workshop experience, weak theory;
good theory, little hands-on practice;
new to both, reads technical texts in the language of instruction slowly.
Shared learning outcome
the apprentice tests a 12 V battery and the charging system with a multimeter and a battery tester, interprets the results and decides what to do, following safety rules.
A three-week unit, 6 × 90-minute lessons, with 16 car mechanic apprentices.
Resources: textbook chapter 7, our school video on battery testing, the tester manual, 4 test benches.
Car mechanic example · Writing the request
Example prompt
You are an experienced automotive VET teacher. Help me design a personalised learning path for a 3-week unit (6 × 90-minute lessons).
Learning outcome: the apprentice tests a 12 V battery and the charging system with a multimeter and a battery tester, interprets the results and decides what to do, following safety rules.
Resources: textbook chapter 7, our school video on battery testing, the tester manual, 4 test benches.
For each profile, propose activities, practice tasks and two checkpoints with decision rules (support loop / standard / extension). All routes lead to the same final practical task, and learners can switch routes. Use only the listed resources and mark anything new. Answer in a table.
Teacher decision · Evaluate and adjust
The draft arrives in seconds and is a good starting point – but the teacher’s review changes several things. The safety steps for disconnecting the battery are incomplete, so the teacher corrects them. A suggested online simulation does not exist and is removed. Profile C gets an illustrated glossary and a video with subtitles instead of a long text. The extension route receives a real fault-finding case from a partner company. Finally, the teacher makes sure the routes are not fixed: every learner can move to a different route after each checkpoint.
| In the draft | Teacher decision |
|---|---|
| Incomplete safety steps | The teacher completes and checks them. |
| A non-existent online simulation | It is removed from the learning path. |
| A long text for Profile C | An illustrated glossary and a subtitled video support learning. |
| An extension task | A real fault-finding case from a partner company is added to the route. |
Teacher decision · Six review questions
AI proposes; the teacher decides. Before sharing a path with learners, it is worth checking it against six questions:
Limits · Safety, inclusion and personal data
AI output can be plausible and wrong. Check content, links and safety-critical steps every time.
an early assessment should not permanently determine the level of tasks or support a learner receives. At every checkpoint, review their progress and adjust the route as needed, taking their views into account.
a path that runs entirely on an e-learning platform can isolate learners and hide problems. Keep regular personal contact, group work and teacher judgement.
Learners’ results and needs are personal data if they can be linked to an identifiable learner. Health information is a special category of data; a description of support needs may also reveal such information.
Use fictional or truly anonymised profiles in general AI chatbots. Labels such as “Profile A” or “Learner 3” do not anonymise the information on their own: if a person can be re-identified, it remains personal data.
Process identifiable learner data only on school-approved platforms, with an appropriate legal basis, a clear purpose, and in accordance with the school’s data protection rules and the necessary data processing agreements. Use only the information genuinely needed: this is data minimisation.
Further information: European Data Protection Board – anonymisation and pseudonymisation (opens in a new tab); European Commission – personal data protection (opens in a new tab).
Takeaways · Flexible paths and teacher responsibility
When designing a personalised learning pathway, the learning outcome remains the same, while the steps, support and pace are adapted to learners’ needs. Drawing on our own trusted resources, AI can help plan activities, learning materials, practice tasks and criteria for moving on. When using AI, we describe learners only through characteristics relevant to their learning, without including names or sensitive data. As teachers, we review, revise where necessary and approve the proposed pathways before use, checking that they align with the learning objectives, are accurate in their subject content, draw on reliable sources and are accessible, practical and understandable for learners. Pathways should remain flexible throughout: at every checkpoint, it should be possible to adjust the next steps, and learners should understand why they are receiving a particular task or form of support.
Adapt the steps, support and pace to learners’ needs.
Activities and practice tasks should be based on checked materials.
Allow learners to change routes and receive different support after checkpoints.
Personal contact, joint workshop tasks and teacher presence remain part of learning.
Self-check quiz
Choose one answer for each question, then check your results. Each answer comes with a short explanation.
Sources · Related materials
The illustrations come from the visual materials supplied with the learning content.
VITA · Contributors and attachments
This work may be reused and adapted under the terms of the Creative Commons Attribution-ShareAlike 4.0 International License (CC BY-SA 4.0).
Funded by the European Union. Views and opinions expressed are those of the author(s) only and do not necessarily reflect those of the European Union or the National Agency.