Using AI tools for T Level lessons at Southport College

KAREN AHMED, LECTURER IN COMPUTING & IT, SOUTHPORT COLLEGE, UK

Introduction

Artificial intelligence (AI) has long been a transformative force in many industries. The launch of ChatGPT in November 2022 sparked a high level of interest in its use in education. This milestone triggered the debate on the benefits and risks of using AI in this context. There are still significant gaps in understanding how to use AI tools in teaching, and existing research often overlooks the current AI skill levels of both teachers and students as well as teaching expertise in this area. This situation leaves teachers without the necessary preparation to use AI effectively. 

Literature review

Celik et al. (2022) provide a review of how teachers use AI. The study highlights both the potential and the challenges posed by the tools, including the gap between AI developers’ skills and their lack of knowledge of effective teaching methods. This hinders the creation of AI tools that align with teaching needs. Hashem et al. (2023) address this gap by examining how teachers can use ChatGPT to reduce their workload. Their findings suggest that automating routine tasks such as lesson planning can free teachers’ time. Building on this idea, Seufert et al. (2021) advocate for augmentation strategies where AI complements – but will not replace – human teachers, for example by using AI to outline lesson plans based on different pedagogical frameworks. Teachers can then concentrate on tailoring lessons to their students’ needs and interests.

Together, these and other studies suggest that using AI in education requires robust tools and skilled teachers who can use them effectively. By treating AI as another tool in their toolkit, teachers can unlock its full potential to enhance teaching and learning. This article demonstrates how ChatGPT and other AI tools can help teachers with lesson planning. AI tools can act as a tool to speed up the process of lesson planning. Tools like ChatGPT can be prompted to provide a template lesson plan to work from, which can then be adapted. This saves time compared to starting a plan from scratch (as is the case with T Levels, where we are often presented with content that we have not taught before).

Can ChatGPT-4o develop a useful T Level lesson plan based on a given pedagogical approach?

Lesson plan 1: Gagné’s conditions of learning theory

Prompt:

Can you explain Gagné’s conditions of learning theory?

Response summary

Gagné’s Conditions of Learning Theory, developed by Robert Gagné, provides a framework for understanding how different types of learning occur. It outlines the conditions necessary for effective learning and is widely used in instructional design. Gagné identified five categories of learning and nine instructional events that support these learning processes.

The response further elaborated on two aspects. One was the five categories of learning: 

• verbal information

• intellectual skills

• cognitive strategies

• motor skills

• attitudes.

The second was the nine events of instruction:

• Gain attention 

• Inform learners of objectives

• Simulate recall of prior learning 

• Present the content

• Provide learning guidance

• Provide feedback

• Assess performance

• Enhance retention

• TransferThe processes of applying learning to new situations.

ChatGPT has been programmed to ask questions to aid the user with writing effective prompts. The question posed was: ‘Would you like an example of a lesson structured around Gagné’s nine instructional events?’ I responded with the prompt of ‘Yes, please’, which resulted in the following:

Response summary

Objective: Students will learn to create flowcharts that represent problem-solving processes using correct conventions, logical structure and sequence.

Key steps:

• Gain attention: Start with a real-world example (e.g. self-checkout machine) to spark curiosity

• Inform objectives: Clearly state the goal: to design effective flowcharts for problem-solving

• Stimulate recall: Connect to prior knowledge of algorithms or step-by-step processes

• Present content: Explain flowchart symbols, structure and conventions with examples

• Provide guidance: Share tips for logical decomposition and common pitfalls to avoid

• Elicit performance: Have students practise by creating a flowchart for a simple task (e.g. withdrawing cash from an ATM)

• Provide feedback: Offer constructive, real-time feedback during the activity

• Assess performance: Assign an individual task to create a flowchart for a more complex scenario (e.g. booking tickets online)

• Enhance retention and transfer: Discuss real-world applications and assign homework to create a flowchart for a household task.

This structured approach ensures that students understand flowchart concepts and apply them to practical problems effectively.

Evaluation

When I’m using these lesson plans in class, I usually change the examples. For this one, I changed the ATM to purchasing a loot box in an online game, where you can use real money or game credits, to make it more relevant for 16-year-olds. I also added some formative assessments, such as questions or quizzes, to check for understanding. I change ‘homework’ to independent study.

Other uses for ChatGPT

I have used ChatGPT for a wide range of things. It has been very useful in delivering the T Level digital production, design and development course for the following activities:

• debugging student code (for example, asking ChatGPT to check for spelling and syntax errors and pasting the section of code to be checked) the student typed the prompt ‘Can you identify the spelling and syntax errors in my code?’ Students acknowledge their use of AI by pasting the prompt and the response into a Word document along with the date on which the tool was used. This is submitted as an appendix along with the work.
• creating business case studies in the style of the assessments
• explaining searching and sorting algorithms and computational thinking techniques
• creating sample test data for database applications
• providing comprehensive feedback on students’ answers to ‘evaluate’ exam questions.

Although it has its uses, ChatGPT is still quite limited in what can be achieved. For example, it does not produce accurate diagrams or flowcharts. 

How can AI tools create T Level resources?

Beyond lesson planning, I have found AI tools useful for generating resources, including reading materials, presentations, worksheets and plenary activities. ChatGPT can do some of these tasks. However, other specialised tools can be more efficient. 

Creating reading materials with Gemini and Hemingway

Gemini offers customised responses and benefits from web access, making it ideal for current affairs. Teachers can use bullet points from a specification to generate content, but they must manually check the accuracy of the outputs. Teachers can use other tools, like Hemingway, to enhance readability and adjust content for different reading ages.

Creating presentations with Gamma

Gamma automatically generates presentations based on lesson notes. The tool creates summarised PowerPoint slides, saving significant time and reducing workload. The free version is limited in the number of slides that it can make, but it is useful for testing before committing to a subscription.

Creating worksheets and activities with TeacherMatic

TeacherMatic is a (paid-for) web app designed for teachers across all levels, from primary to higher education. It enables the creation of various student activities for use in lessons. These include resources like multiple-choice questions, fill-in-the-blank exercises and T Level projects. Resources can also be based on uploaded documents, such as Gemini-generated articles.

Creating SEND resources with Notebook LM

Notebook LM is another free Google tool. It can be used to create a podcast from a written PDF document. Dyslexic students on the T Level have found it useful to have the document spoken to them and they felt that it improved their understanding of the content. The tool offers only two chatty American voices, but the students found the ‘podcast’ style engaging. However, some students did feel that the voices were too robotic.

Creating project plans and rubrics with MagicSchool

MagicSchool offers various tools for teachers and students. Teachers can create project plans for project-based learning, complete with rubrics. While the initial output needed adjustments for the T Level, it provided a starting point for a student research project. The generated rubric was adapted into a peer assessment and feedback activity. Using these tools made the process much quicker and easier than writing everything from scratch.

Conclusion: What do teachers need?

AI tools greatly reduce the workload when designing lessons from scratch. They offer structure and content suggestions. However, there are still limitations. AI’s datasets are vast but lack depth in teaching methods and classroom applications. These aspects are improving, but full acceptance and integration will take time.

Many teachers face challenges in adopting AI tools due to limited digital confidence and understanding of the technology. For AI to be truly effective in education, ongoing training for teachers is essential.

The examples of AI use and specific tools in this article are for context only. They do not imply endorsement or recommendation of any particular tool or approach by the Department for EducationThe ministerial department responsible for children’s services and education in England or the Chartered College of Teaching and any views stated are those of the individual. Any use of AI also needs to be carefully planned, and what is appropriate in one setting may not be elsewhere. You should always follow the DfE’s Generative AI In Education policy position and product safety expectations in addition to aligning any AI use with the DfE’s latest Keeping Children Safe in Education guidance. You can also find teacher and leader toolkits on gov.uk .