Adapting the ADDIE model to use AI

Contents:

• Key aspects of the ADGIE model
• Effective design of the educational process using the ADGIE model
• The evolution of the role of the instructional designer in the era of artificial intelligence

Every day, artificial intelligence tools are becoming more and more popular in the process of creating training courses and materials. However, at the moment there is no unified and systematic approach to their use, which could help minimize risks and increase the effectiveness of neural networks. At the same time, a number of developments related to this issue already exist.

Previously, we discussed the GAIDE framework, created by American scientists for the effective generation of educational materials. Now, Moroccan researchers Khadija Hilali and Meriem Chergui have proposed the idea of ​​integrating neural network technologies directly into ADDIE, which is the most widely used model for developing educational courses.

The classic ADDIE model does not take into account human interaction with artificial intelligence and the opportunities for individualization of the learning process offered by neural networks. Following a survey of education professionals, the researchers developed their own adaptation of the ADDIE model for the use of AI technologies, which was named ADGIE.

In this article, we will consider the key characteristics of the adapted model.

Key Aspects of the ADGIE Model

Khadijah Hilali and Meryem Chergui sought to integrate elements of the ADDIE model with the potential of artificial intelligence to meet the modern requirements of education professionals. In developing their model, they relied on data obtained from a survey of 90 professionals, including trainers, teachers, and curriculum designers, who expressed their expectations and needs regarding the use of AI in their work.

Read also:

Creating training courses using artificial intelligence: what advantages and disadvantages have been discovered Based on the findings from the survey, the researchers created the ADGIE model. The model's name, similar to the classic ADDIE model, is an acronym composed of the initial letters of its components. The main difference from ADDIE is that ADGIE uses generation and individualization stages instead of development and implementation stages.

Efficient design of the educational process using the ADGIE model

The updated model uses artificial intelligence tools while maintaining human supervision.

Similar to the classic ADDIE model, at this stage, an analysis of the needs of both the customer of the educational course and the market as a whole is carried out, and research of the target audience is also conducted.

The tasks of artificial intelligence involve generating questions that will help explore the target audience of the course: students' knowledge level, their expectations of the educational process, their needs, and interests. After this, it is necessary to analyze the responses collected through surveys or interviews and, based on the data obtained, create generalized images of various audience segments, which in product analysis are usually called personas.

The responsibilities of the instructional designer include ensuring that the characteristics of the personas created by artificial intelligence correspond to the identified educational needs, as well as the goals and objectives of the course. This will form a solid foundation for subsequent stages of development.

At this stage, the basis of the course is formed, the order of modules and lessons is established - in essence, a kind of "skeleton" of the curriculum is created.

The main responsibilities of the instructional designer are to select high-quality, modern and relevant learning resources that will form the basis for course development.

The tasks of artificial intelligence are to organize the collected information - to form the thematic structure of the course and the student learning path map (SJM), to develop a rough plan or scenario, and to create a prototype of the educational program.

In the traditional ADDIE model, this stage is called "Creation", but its essence remains unchanged - it is the process of preparing all the learning resources that students will work with.

Artificial intelligence functions: after the instructional designer approves the course, AI begins to create its content - this includes preliminary versions of learning materials, graphic images and diagrams, assignments and exercises, as well as test questions and much more another.

The responsibilities of an instructional designer include carefully reviewing the created learning materials, analyzing their quality and compliance with the stated educational goals, as well as the needs of students. Based on this assessment, they must make any necessary adjustments independently.

The fourth stage of the classic ADDIE model, when students begin their learning within the course, is called "Implementation." The updated version of this model emphasizes how to use artificial intelligence to adapt the educational process.

During the learning process, artificial intelligence analyzes data on student behavior and their progress in real time, which allows it to modify learning materials and learning paths. For example, if a student is having difficulty with an oral explanation in a video lecture, a neural network can create a text summary. Additionally, if a student expresses interest in a specific subtopic, the system can provide additional resources for deeper learning.

The instructional designer's responsibilities include monitoring student progress and achievements, as well as ensuring that the changes and advice provided by the artificial intelligence are truly relevant and effective. The main goal is that these recommendations contribute to the achievement of students' learning goals.

In the traditional ADDIE approach, the assessment stage is the final one. At this stage, the instructional designer examines course performance and collects feedback from students and the client to identify areas for improvement and make adjustments accordingly. In contrast, the developers of the ADGIE model implemented a methodology that involves evaluation at each stage:

• In the process of analysis, the instructional designer evaluates how accurately and reliably the learner personas have been developed, and also checks the correctness of the formulation of the needs of the target group and the learning objectives.
• In the process of developing instructional design, it is important for the specialist to make sure that the organization of the program meets both the educational goals and the requirements of the target audience.
• In the process of developing educational content, the instructional designer monitors its quality.
• In the process of individualization of learning, students analyze their educational path and provide their assessment.

The evolution of the role of the instructional designer in the age of artificial intelligence

Philippa Hardman, formerly a research fellow at the University of Cambridge and developer of the DOMS™️ learning design system, recently expressed her opinion on how artificial intelligence is transforming the tasks facing instructional designers.

According to the researcher, The results of a survey conducted to form ADGIE indicate a shift in the perception of the role of instructional designers and methodologists in the development of educational programs. Hardman identified three key areas of these changes and offered recommendations for adapting to them.

According to the survey results, the majority of trainers, teachers, and instructional design specialists (94%) expressed a desire to receive a variety of learning materials from artificial intelligence, including exercises and other types of interactive content. At the same time, 91% of study participants believe that human control over the quality of created content should remain with humans. Thus, as Philippa Hardman notes, neural networks will be able to develop drafts of educational materials, while humans will be engaged in selecting the best options and further refining them.

How to prepare:

• Improve your skills in selecting materials, as well as in analyzing their real-world accuracy, reliability, relevance, and relevance to the topic.
• Create a personal list of criteria that will allow you to evaluate content created by artificial intelligence based on key parameters.

Despite the fact that 83% of professionals surveyed expressed a willingness to use AI support in curriculum development, only 49% of them would be willing to trust a neural network to make decisions regarding instructional methods and pedagogical practices. Hardman emphasizes that this indicates that responsibility for learning strategy still lies with the instructional designer, who makes important methodological decisions. Based on a specific strategy, it formulates tasks for the AI, which in this case acts as the executor.

How to prepare:

• Enriching and deepening your knowledge in the field of education, as well as exploring related sciences and modern theories is an important task. Hardman argues that understanding the processes people go through to learn is the most important source of information for an instructional designer.
• Study prompt engineering to transform your educational ideas into clear directions for neural networks.

Read Also:

Competition: Who will create a better course—an instructional designer using a neural network, or the traditional way, without its help?

As the level of automation increases, according to Philippa Hardman, so does the responsibility of the instructional designer regarding the ethical issues associated with the systems they create.

How to prepare:

• Deepening knowledge in the field of artificial intelligence ethics includes studying how algorithms can reflect biases or allow the misuse of personal data. This will allow us to more effectively identify and address such shortcomings.
• Improving data analytics skills is necessary to ensure that the implementation of artificial intelligence in the educational process leads to positive outcomes for all students.

Read also:

• Five skills that are changing the work of instructional designers in light of the introduction of neural network technologies.
• 8 ways to use the Perplexity neural network for teachers and methodologists.
• When creating an online course, ChatGPT has both advantages and disadvantages.

  Among its strengths are the ability to generate a variety of content and adapt to different topics and styles. The model can quickly create learning materials, assignments, and tests, which significantly saves the developer's time. It is also worth noting its ability to suggest ideas for structuring the course and maintain interaction with students, answering their questions and providing explanations.

  However, ChatGPT also has its limitations. For example, the model may not always accurately understand the context or nuances of the topic, which can lead to the creation of lower-quality content. In addition, the lack of deep knowledge in specific areas can affect the accuracy of the information it provides. It is also important to consider that, despite its ability to facilitate dialogue, ChatGPT cannot replace personal contact and the emotional connection between instructor and students.

  Therefore, when using ChatGPT to develop an online course, it is necessary to consider both its capabilities and limitations in order to integrate this tool into the educational process as effectively as possible.

• Creating Online Courses Using Neural Networks: Case Studies
• The IDEAS Framework for Distance Learning: What It Is and How It Works.
• The time has come to discuss ethical issues in the EdTech sector. With the development of technology in the educational field, new challenges and responsibilities arise that require attention. The emergence of innovative solutions such as artificial intelligence and data analytics opens up many opportunities to improve learning, but along with this come risks associated with privacy, access to information, and data manipulation.

  It is necessary to understand that technology can both enrich the educational process and create potential threats. In this regard, it is important to consider how ethical standards can be integrated into educational technology practices. Discussing these issues will help create a safer and more inclusive educational environment that takes into account the interests of all stakeholders. The need for an ethical approach is also linked to the responsibility of developers and educational institutions, who must be prepared to consider the impact of their decisions on students and teachers. It is important not only to monitor compliance with existing standards but also to actively work to develop new standards and principles that reflect modern realities and challenges.

  So, the EdTech community must openly discuss ethical issues to ensure that technology serves the benefit of education and does not become a source of new problems.