Instructional design models for the use of artificial intelligence tools and systems

1. Introduction

Instructional (learning) design models provide structured frameworks and guidelines that help teachers and educational designers systematically plan, develop, and evaluate educational activities. With the advancement of artificial intelligence, these models gain new capabilities – they enable more precise recognition of student needs, adaptation of content to different learning styles, and more effective monitoring and evaluation of progress.

One of the most well-known and widely used models in educational practice is the ADDIE model, which can be easily supplemented with modern tools and approaches based on artificial intelligence. Combined with artificial intelligence, this model becomes a powerful framework for developing personalised, motivating, and data-driven educational experiences.

Below are the characteristics of selected instructional design models that can assist educational stakeholders in introducing artificial intelligence in education.

2. Topic: ADDIE model with integration of artificial intelligence tools

ADDIE is the most widely used instructional design model and consists of five phases: Analysis, Design, Development, Implementation, and Evaluation (Figure 1).

Each stage can be significantly enhanced by applying artificial intelligence.

Integrating artificial intelligence into the ADDIE model enables more precise analyses, personalised learning design, faster creation of educational materials, dynamic implementation, and continuous evaluation. The following explains how AI can advance each phase of educational design.

Additional articles:

• The Role of Artificial Intelligence In Instructional Design
• The Role of AI In The Creation and Deployment of eLearning With ADDIE 

3. Examples of other instructional design models and applications of artificial intelligence

In addition to the ADDIE model, other learning design models that support the effective integration of artificial intelligence (AI) tools and systems are increasingly used in educational practice. These models offer greater flexibility, faster iterative cycles, and better adaptation to users’ real needs, making them especially suitable for dynamic, data-driven educational environments.

Traditional instructional design models typically follow a linear, sequential process in which content is planned, developed, and evaluated in advance, often without the ability to make quick adjustments during course implementation. In contrast, AI-assisted approaches emphasise iterativity, adaptability, and learning analytics, enabling continuous improvement and personalisation of the educational experience.

It is important to note that artificial intelligence does not operate independently or replace the instructor or learning designer. Its role is to support professional decisions and pedagogical judgment, for example, through data analysis, content suggestions, or automation of routine tasks. Human expertise remains central to the process, ensuring the relevance, ethics, and quality of educational solutions.

These approaches shift the focus from assumption-based design to data-driven design and real-time feedback. In this context, the instructor and instructional designer act as reflective process guides who, with the support of artificial intelligence, monitor students’ needs and adapt content according to their progress.

Among the most commonly used models are:

• The agile model, which emphasises iterative development and constant adaptation to user feedback

• Kemp’s model, which views learning design as a complete system, considering context, goals, content, and student support

• The Successive Approximation Model (SAM), which relies on rapid prototyping and continuous improvement of educational solutions

• The User-Centered Design (UCD) model, which places the student, or user, at the center of planning and creating educational content.

Below, you will study the principles of each of these models in more detail, along with examples of their application in the context of artificial intelligence. This will provide insight into how different learning design approaches can support the development of more effective, personalised, and engaging educational experiences.

While these models are among the most well-known, it is important to note that many other approaches can be adapted to contemporary educational needs and the specific characteristics of individual educational environments.

Artificial intelligence does not replace instructors or instructional designers but helps them make more informed decisions.

The key to successful implementation of AI tools is balancing technological support with human expertise.

Recommended resources for further reading:

• Thinkific: Best ChatGPT Prompts for Instructional Designers
• Devlin Peck: AI in Instructional Design (Tools and Examples)
• CommLab India: Instructional Design Challenges and Possibilities with AI
• eLearning Industry: Role of Artificial Intelligence in Instructional Design
• Articulate: How AI Is Transforming Instructional Design
• University of San Diego: The Role of AI in Instructional Design
• AI-Powered Instructional Design: A Practical Guide To Staying Ahead
• AI In Instructional Design: Exploring The Future of eLearning Content Creation
• Will The Rise Of AI Eliminate Instructional Design By Humans?—Part 1
• 40+ AI Prompts For Instructional Designers: From Storyboards To Scenario-Based Learning

3.1. Agile Instructional Design Model

The agile learning approach originates from software development practices and is based on rapid iterations, constant feedback, and a high level of adaptability. In education, this approach allows content and teaching methods to be continuously improved based on student feedback and learning analytics results.

How artificial intelligence supports the agile model in education:

• Rapid creation and updating of educational materials – Artificial intelligence enables teachers and instructional designers to develop and adapt content quickly. Tools such as ChatGPT can generate texts and questions, Synthesia can create video lessons, and Canva can design visuals.

• Tracking student engagement and understanding – AI-powered analytics allow real-time monitoring of progress and engagement levels. Based on this data, teachers can immediately adjust their approach and content to meet student needs.

• Conducting A/B testing – AI systems can automatically test different versions of the same educational material (such as lessons, quizzes, or videos) to determine which approaches produce better learning outcomes.

Example: 

Instructors regularly adjust their teaching approaches to complex topics. After each lesson, an AI-powered system analyses data on student engagement and performance and suggests ways for teachers to design and adapt the next lesson. In this way, content is continuously improved, and students receive personalised and motivating educational materials aligned with their needs and learning pace.

Advantages and disadvantages of applying the agile model with artificial intelligence:

Advantages: 

• Enables rapid adaptation of content and teaching methods based on student feedback and learning analytics

• Encourages creativity and experimentation in instructional design, as AI tools facilitate testing of different approaches and formats

• Supports continuous improvement of teaching quality through data-driven decisions

• Reduces the burden on teachers for routine tasks (such as creating materials, quizzes, and checking comprehension)

• Contributes to the creation of dynamic, interactive, and personalised educational experiences

Disadvantages:

• Requires constant pedagogical reflection and supervision, as over-reliance on AI can reduce the human aspect and pedagogical warmth of teaching

• Implementation can be challenging for teachers who lack digital competencies or experience with agile methods

• Frequent updates and iterations can lead to content fragmentation if clearly defined learning objectives are not maintained

• Raises questions of ethical responsibility and privacy protection regarding data used in learning analytics

• Requires additional time and team coordination, especially in more complex teaching environments

Reflective questions:

• How could an agile approach be applied in your subject with the support of AI?
• Which activities or phases of teaching could be improved with rapid iterations and adjustments based on learning analytics?

You can read more about the agile model and the application of AI:

Agile Instructional Design Framework

ELM Learning: “Agile Instructional Design: Key to Flexible, Effective Learning.”

Olaide et al. (2024): Artificial Intelligence with Agile Teaching Strategies for Personalized Learning

Image source: Shutterstock

3.2. Successive Approximation Model (SAM)

This model enables the effective integration of artificial intelligence tools at different stages of learning design, accelerating adaptation and increasing precision in the creation of materials. Artificial intelligence assists in prototyping, testing, analysis, and adaptive improvement of educational resources.

How artificial intelligence supports the SAM model:

• Prototyping with artificial intelligence
  With AI tools, sketches, scenarios, and educational content can be quickly created and then tested with learners.

• Testing analysis
  Artificial intelligence analyses data on learner behaviour and engagement during testing and automatically generates reports on the effectiveness of content and activities.

• Adaptive content improvement
  Based on testing results, AI can suggest changes to lesson structure, questions, examples, or visuals to achieve greater clarity and learning effectiveness.

Example:  

When creating an e-course on critical thinking, a team of instructors uses ChatGPT to quickly generate different versions of lesson scenarios. These versions are tested on a small group of students through microtests. AI analyses the results, identifies the parts that cause the most difficulties, and suggests specific changes. In this way, the content is gradually improved, and students participate in creating higher-quality and more relevant educational material.

Advantages and disadvantages of applying the SAM model with AI

Advantages:

• Enables rapid testing and improvement of content before final implementation
• Encourages student involvement in the development process through feedback and microtests
• Enables data-driven decisions about adjustments and changes
• Reduces the risk of failure because problems are identified early in the design phase
• Encourages collaboration between teachers, designers, and AI systems.

Disadvantages:

• Requires more time in the initial stages due to repeated testing
• Can create data saturation if analytics are not clearly structured
• Excessive automation can reduce the creativity and pedagogical intuition of designers
• Requires responsible interpretation of artificial intelligence analysis results and human judgment when making decisions about changes.

Reflective question:

• In your practice, how could you use an iterative approach and artificial intelligence tools to test and improve teaching materials before their final implementation?

More details about the SAM model and its use:

Successive Approximation Model (SAM)

SAM model in E-learning Development

SAM Model: An Agile Approach To Instructional Design

3.3. Kemp Instructional Design Model

The Kemp model views the learning process as an interconnected whole, where goals, content, methods, technology, and participant needs are all linked. It also considers the broader educational context, including organisational, technological, and pedagogical factors that shape the learning experience.

How AI supports the Kemp model:

• Broader context analysis
  Artificial intelligence can analyse institutional needs, teachers’ digital literacy, and technology availability, and suggest optimal approaches and tools for individual educational programs.

• Learning personalisation
  AI systems can recommend content that matches each learner’s prior knowledge, learning style, and goals, thereby increasing engagement and the effectiveness of the learning process.

• Automating and adapting content for different groups
  A single educational module can be quickly adapted for different users – such as students and teachers – by using AI tools to translate, adjust tone, language, or media presentation format.

Example:

The Moodle system uses an AI-based add-on that automatically creates individual lesson and quiz schedules for each student. The system considers each learner’s previous performance, engagement, and learning pace, and personalises the content and schedule of activities. In this way, AI directly supports Kemp’s holistic approach, where planning, development, and evaluation are interconnected and focused on the individual.

Advantages and disadvantages of applying Kemp's model with artificial intelligence.

Advantages:

• Enables a holistic approach to learning design, encompassing all elements of the educational process – goals, content, methods, evaluation, and technology.

• Artificial intelligence facilitates the analysis of complex data on students, teachers, and institutional needs, supporting informed decision-making.

• Encourages personalisation and inclusiveness, as AI can adapt content to different learning styles and paces.

• Enables automated progress monitoring and evaluation, reducing the administrative burden on teachers.

• Contributes to the strategic planning of teaching development through data integration at the institution-wide level.

Disadvantages:

• Implementation requires a high level of coordination and collaboration among teachers, IT specialists, and administration.

• Overreliance on analytical data can lead to neglect of qualitative aspects of learning.

• Implementation can be technically and organisationally complex, especially in larger educational systems.

• Requires ongoing professional development for teachers to effectively use artificial intelligence tools.

• Raises ethical issues regarding privacy protection and the fairness of algorithms in assessment and recommendations.

Reflective question:

• How could you apply Kemp's holistic approach in your teaching, supported by artificial intelligence that connects analytics, personalisation, and evaluation?
  
• Which elements of the educational process would you most like to improve using such an integrated approach?

More about Kemp's model and artificial intelligence support:

Kemp Instructional Design Model

Kemp Design Model: A Flexible Approach to Practical Learning Experiences

Applying The Kemp Design Model In eLearning: A Guide For eLearning Professionals

3.4. User-Centered Design (UCD) Model

How AI supports the UCD model:

• Monitoring user behaviour 
  AI analyzes content usage patterns, time spent on specific activities, and methods of interacting with materials. Based on this data, the system suggests personalised adjustments to enhance the learning experience and increase student engagement.

• Chatbots and virtual assistants 
  AI-powered chatbots provide ongoing support to students by answering questions, explaining concepts, and offering additional resources when the system detects uncertainty or errors. Beyond their informative role, these tools also serve a motivational role, helping students maintain continuity and confidence throughout the learning process.

• Empathy analytics 
  Advanced AI systems analyse the tone, rhythm, and language of learners’ communication and can recognise emotional indicators such as frustration, demotivation, or confusion. When these are detected, the system automatically offers further clarification, suggests activities, or provides messages of support, fostering a more empathetic and human-centered digital learning environment.

Example:

An online course for adults uses an AI system that dynamically adjusts learning paths based on learners’ behaviour and reactions. For those who progress quickly, the system offers shorter modules and advanced tasks, while for those experiencing difficulties, it provides additional explanations, visuals, and extended time for tasks. In this way, AI fulfills the fundamental principle of the UCD approach – focusing the educational experience on the individual needs of each user.

Advantages and disadvantages of applying the UCD model with AI.

Advantages:

• Places learners at the center of the educational process, increasing engagement, motivation, and a sense of control over learning.
• Enables a high degree of personalisation through behavioural analytics and adaptive AI recommendations.
• Encourages an empathetic and responsive approach in the online environment, as AI systems can recognise emotional signals and respond promptly.
• Supports continuous improvement of content based on feedback and real user experiences.
• Relieves teachers by providing automated support and micro-interventions during the learning process.

Disadvantages:

• Requires the collection and processing of large amounts of personal data, raising concerns about privacy, ethics, and algorithm transparency.
• May lead to excessive automation in communication, diminishing the authenticity and human warmth of teacher support.
• Carries a risk of misinterpreting emotional signals, especially in multilingual or culturally diverse groups.
• Implementation requires technical support and ongoing monitoring of the AI solution’s effectiveness.
• Over-reliance on analytics can result in a superficial understanding of complex educational experiences.

Reflective question:

• How could you apply the principles of user-centered design in your teaching with the support of artificial intelligence, while still maintaining a human dimension in communication and student support?

Basics and application of UCD:

What is User Centered Design (UCD)?

User-Centered Design For eLearning

4. The importance of pedagogical methods in modeling teaching through instructional design models and the application of artificial intelligence

Pedagogical methods are the foundation of every teaching and learning process, as they determine how teaching will be modeled – how the learning process will be structured, guided, and evaluated.

Without a clear understanding and consistent application of appropriate pedagogical methods, the use of instructional design models and artificial intelligence tools can become ineffective or superficial. Pedagogical methods serve as a bridge between theoretical models and practical teaching, enabling technology, including artificial intelligence, to support the achievement of learning goals rather than being an end in itself.

By applying pedagogical methods, we can:

• Identify educational approaches that best suit the needs, knowledge level, and motivation of participants.
• Define activities and forms of interaction that foster meaningful and engaged learning.
• Guide content design, evaluation, and feedback toward actual learning outcomes.
• Select methods that enable the integration of artificial intelligence in ways that enhance engagement, reflection, and understanding.

In other words, pedagogical methods provide a framework that shapes how instructional design models are implemented in practice. In this context, artificial intelligence serves as a support and extension of pedagogical practice, not as a replacement.

4.1. Types of pedagogical methods for applying artificial intelligence (AI) tools and systems

When applying artificial intelligence in education, it is essential to select pedagogical methods that align with technological capabilities and educational objectives. Artificial intelligence does not replace the pedagogical process but serves as a tool to enhance, strengthen, and increase its efficiency. In this context, some of the most common and effective methods through which AI can significantly contribute to the quality of teaching include the following:

By carefully selecting pedagogical methods and integrating them with artificial intelligence technology, instructional designers can create dynamic, customised, and effective educational content that fully leverages the capabilities of artificial intelligence.

4.2. Example: Socratic dialogue and the application of artificial intelligence

Socratic dialogue is a fundamental pedagogical method that develops critical thinking and deeper understanding through questions and answers. In modern education, artificial intelligence (AI), especially large language models (LLMs) such as ChatGPT, significantly enhances this method.

How does AI support Socratic dialogue?

AI-powered chatbots and virtual tutors can take on the role of “Socrates”, asking thoughtful questions that encourage reflection. These systems analyse responses and adapt further questions according to students’ understanding, offering personalised learning. LLMs further automate questioning, adapt the dialogue, analyse answers, and provide feedback, simulating complex philosophical conversations.

Advantages of the application:

• Learning is personalised through guided discussions.
• Encourages active learning and self-reflection.
• Accessible online, with AI moderating the dialogue in real time or asynchronously.
• Increases engagement and develops metacognitive skills.

The integration of Socratic dialogue with artificial intelligence tools, especially LLMs, creates a powerful synergy of tradition and technology. It significantly enhances the quality and depth of the educational process, offering personalisation and scalability in learning.

5. Important considerations in using AI in instructional design

When introducing artificial intelligence (AI) into education, it is important to recognise that great opportunities come with significant responsibilities.

Key points to consider:

Data Privacy – The Foundation of Trust 

AI systems in education often collect and analyse large amounts of student data. It is essential to protect the privacy and security of this data. Instructors and teachers must comply with laws and regulations such as the GDPR. Transparency in data collection and usage is not only a legal obligation but also the foundation for maintaining the trust of parents and students. Without trust, the successful implementation of AI is not possible.

Bias in artificial intelligence models – A challenge for equitable education

AI systems learn from the data on which they are trained. If this data contains biases, AI can inherit them, leading to unfair recommendations, discrimination, or inaccurate student evaluations. It is our responsibility to continuously monitor and adjust AI models to ensure they are fair and inclusive for all learners, regardless of background.

Preserving the Human Factor in Education  

Although artificial intelligence is taking over many routine tasks – from data analysis to content generation – the human element remains essential for meaningful and high-quality learning.

Instructors are no longer just transmitters of knowledge; they are designers of the learning experience. They recognise context, select appropriate pedagogical strategies, and provide support that cannot be replaced by AI.

AI offers personalisation and efficiency, but it cannot replace relationships, empathy, or ethical reflection. Instructors guide participants in using AI tools, foster critical literacy, and ensure active, reflective learning.

In conclusion, the student remains at the center, and collaboration between humans and AI creates opportunities for better, more humane education.

Image source: Shutterstock

6. Conclusion and the future of instructional design with AI

Artificial intelligence will continue to transform how we learn and teach, enabling greater personalisation, adaptability, and engagement. The use of virtual assistants, automated evaluation systems, and immersive technologies – such as virtual reality (VR), augmented reality (AR), and mixed reality (MR)—along with artificial intelligence, is expected to become increasingly widespread.

Potential for Further Development and the Role of the Instructional Designer

The role of instructional designers is shifting from technical and operational tasks to the strategic design of learning experiences. Designers are no longer just content planners but are now architects of interactive and personalised learning, leveraging artificial intelligence to:

• Create content in collaboration with AI tools and systems.
• Analyse learning data and adapt materials.
• Monitor participant engagement and progress in real time.
• Shape new forms of experiences, such as gamification, simulations, and VR/AR environments.

Additionally, ongoing development in new technologies, pedagogical methods, and ethical considerations is essential for maintaining the quality and relevance of education.

Image source: Shutterstock

7. Literature