The Post-AI Instructional Designer
How the ID role is changing, and what this means for your key skills, roles & responsibilities
Hey folks! 👋
An increasing amount of research shows that optimal workplace performance happens when we combine the powers of expert human and AI.
The big question I’ve been exploring this week is: what should a partnership of human instructional designer and AI look like in practice, and what does this mean for our key skills, roles and responsibilities?
The Delegation Model of Human-AI Interaction
According to a lot of research published in the last few years, there are a number of tasks which are “innately human” and some that are best delegated to AI.
Here’s the TLDR on what this research found:
AI’s Strengths & Role
Automation and Repetitive Tasks: AI is particularly effective at handling repetitive tasks, data processing, and tasks that require speed and precision. This includes tasks such as data entry, basic decision-making, and some forms of content generation (Kim et al., 2020).
Decision-Making and Predictive Analytics: AI excels in tasks that involve predictive analytics and decision-making based on large datasets, often surpassing human capabilities in these areas when it comes to speed and accuracy (Raghu et al., 2019).
AI Delegation and Task Management: AI can also manage certain tasks by deciding whether to perform a task itself or delegate it to a human, optimising the performance of human-AI teams (Hemmer et al., 2023).
Human’s Strengths & Role
Complex Decision-Making and Creativity: Humans are better at tasks that require complex decision-making, creativity, and emotional intelligence. These include strategic decision-making, innovation, and tasks that require a deep understanding of context and human needs (Gul, 2023).
Building Trust and Ethical Considerations: Humans are essential in roles that require building trust, particularly in management and leadership positions. While AI can perform evaluations, human managers are often perceived as more trustworthy in certain contexts, especially when fairness and ethical considerations are crucial (Qin et al., 2023).
Collaboration with AI: In tasks where human judgment and AI capabilities complement each other, such as effective communication with AI (prompting) & collaborative decision-making, humans play a crucial role in overseeing and guiding AI processes to ensure optimal outcomes (Luong et al., 2020).
According to this research, humans and AI have distinct roles. As a result, the relationship between AI and instructional designer is one of task separation and delegation: the human does what they’re good at, and the AI does what it’s good at.
The Delegation Model & Instructional Design
The Delegation Model of AI-human collaboration is increasingly common among instructional designers I work with on the ground.
In practice, the typical human AI-ID dynamic currently follows the Delegation Model described by researchers, and looks something like this:
AI’s Role in Instructional Design
Content Creation: IDs delegate a variety of time-consuming content generation tasks to AI, from email-writing to quiz creation, image generation and “text to video” production.
Information Retrieval & Synthesis: IDs lean into AI’s ability to search through vast amounts of information and synthesise it into coherent summaries. Many IDs use AI to help them to make sense of large quantities of complex information provided by subject matter experts at the start of their process, or to compile and summarise literature reviews.
Translation: IDs lean into AI’s ability to handle structured tasks that require speed and precision - including content translations.
Data-Analysis: some IDs use AI to process and analyse large amounts of data, e.g. analysing learner performance data or generating reports on course completion rates.
Human’s Role in Instructional Design
Creativity & Innovation: Generating ideas, especially those that connect seemingly unrelated concepts, is considered to be a complex skill best tackled by humans.
Effective & Empathetic Communication: Communicating in ways which are clear, “humanised” and tailored to a specific audience is considered to be an innately human skill.
Strategic Thinking: Developing long-term visions and strategies that consider a multitude of factors, including organisational culture and industry trends, is considered to be most effective when completed by humans.
Ethical Decision-Making: Navigating complex ethical dilemmas, e.g. in learning contexts with diverse audiences, is considered to require purely human judgment and values
Research shows that this “Delegation” model enables instructional designers to make some significant gains in efficiency (speed) and smaller but still measurable gains in effectiveness (quality of output).
However, according to a growing, parallel body of research - including a study from the University of Michigan published just last week - the most effective model of human-AI interaction (i.e. that which leads to most gains in both effectiveness and efficiency) is more nuanced and complex than simply dividing tasks between human and AI.
The Partnership Model of Human-AI Interaction
According to a growing body of research, including a study from the University of Michigan published last week, the line between the capabilities of human and machine may be more blurred than the Delegation Model suggests.
In an experiment involving school teachers, researchers found that optimal performance (efficiency and effectiveness) occurred when teachers used AI not for isolated task completion but for functional, creative and strategic input on their end to end workflow.
Specifically, the study revealed that teachers who reported most productivity gains were those who used AI not just for creating outputs (like quizzes or worksheets) but also for seeking input on their ideas, decisions and strategies.
Those who engaged with AI as a thought partner throughout their workflow, using it to generate ideas, define problems, refine approaches, develop strategies and gain confidence in their decisions gained significantly more from their collaboration with AI than those who only delegated functional tasks to AI.
TLDR: research increasingly suggests that the most productive human-AI interactions are those which take a more integrated approach than simply delegating functional “output tasks” to AI.
In the Partnership Model, AI serves as an "always-on" thought partner, contributing to both strategic planning and task execution throughout the worker’s entire end to end workflow.
The Partnership Model & Instructional Design
So what would this model like in practice for instructional designers?
For illustrative purposes, let’s think about what the Partnership Model looks like in the context of ADDIE:
Analysis:
AI: Analyses data, suggests potential learning objectives, provides insights on learner characteristics
Human: Collaborates with AI to define problems, conducts stakeholder interviews, refines AI-suggested objectives
Design:
AI: Proposes instructional strategies, generates ideas for learning activities
Human: Engages in dialogue with AI to refine strategies, combines AI suggestions with personal expertise
Development:
AI: Generates content drafts, suggests interactive elements, assists in storyboarding
Human: Co-creates with AI, iteratively refining content and incorporating AI suggestions
Implementation:
AI: Provides real-time feedback on engagement, suggests adaptive content changes
Human: Collaborates with AI to make real-time adjustments, uses AI insights for continuous improvement
Evaluation:
AI: Provides predictive analytics, suggests potential improvements
Human: Engages in dialogue with AI to interpret data, collaboratively develops improvement strategies
The Partnership Model requires a more sophisticated set of skills that go beyond simply using AI tools to generate outputs.
In order to feel the optimal benefits of AI for both our efficiency and effectiveness, instructional designers must more deeply integrate AI into their end to end thought process and workflow. This has some interesting implications for what we consider to be the key skills of an effective instructional designer.
The Post-AI Instructional Designer
So what does this all mean for the role and skills of the instructional designer?
Traditionally, the role and key skills of the ID goes something like this:
Needs Analysis: the ability to conduct thorough needs analyses and design appropriate assessments was crucial. IDs needed to identify performance gaps, understand learner characteristics, and create assessments that accurately measured learning outcomes.
Understand Instructional Design Theories and Models: a deep understanding of learning theories, instructional models and pedagogical approaches is essential for creating effective learning experiences. IDs needed to apply these theories to design courses that facilitated learning and knowledge retention.
Content Development & Authoring: as IDs, we are often responsible for creating content, writing learning materials, and developing multimedia assets. This requires strong writing skills, creativity, and proficiency in various authoring tools and multimedia software.
Project Management and Stakeholder Collaboration: a key part of the ID role is managing the entire course development process, which requires strong project management skills. We also needed to collaborate effectively with subject matter experts, production teams, and other stakeholders to bring learning projects to completion on budget and on time.
In the post-AI world, this role doesn’t change, but the way that we execute it (and the skillset required to to optimise for efficiency and effectiveness) does:
Needs Analysis: IDs still need to identify performance gaps and understand learner characteristics, but AI now plays a significant role in this process. IDs must be skilled in using AI-driven analytics tools to gather and interpret large datasets on learner behaviour and performance. The focus shifts from manual data collection to AI literacy and data-informed decision making. IDs need to critically evaluate AI-generated insights, combining them with human intuition to design more targeted and personalised learning experiences.
Understand Instructional Design Theories and Models: the fundamental theories remain important, but the skill of prompt engineering and deep knowledge of instructional design theory becomes more crucial here, as IDs need to effectively communicate instructional design principles to AI tools and assess the quality of AI’s suggestions.
Content Development & Authoring: While content creation is still a key responsibility, IDs now often act as curators and refiners of AI-generated content rather than creating everything from scratch. The focus shifts to prompt engineering skills to guide AI in generating initial content, and then applying critical analysis, evaluation and refinement skills to ensure the AI-generated material aligns with learning objectives. Creativity is still crucial, but it's enhanced by AI’s input and applied more to enhancing and contextualising AI-generated content.
Project Management and Stakeholder Collaboration: In the post-AI world, project management involves orchestrating a complex interplay of human and AI resources. IDs need skills in AI-human workflow management, understanding how to effectively allocate tasks between AI tools and human team members. Collaboration now extends to "working with" AI tools, requiring skills in prompt engineering and iterative co-creation. Stakeholder management becomes more complex, as IDs need to address concerns about AI integration, manage expectations, and communicate the value and limitations of AI in the learning design process.
In all of this change, the ID’s role is elevated rather than eliminated. Among the most critical skills for the post-AI ID are:
Advanced Knowledge of Learning Science & Instructional Design Research:
Possessing a deep, nuanced understanding of learning theories and instructional models
Using theoretical knowledge to prompt, critically evaluate and guide AI-generated instructional strategies
Data Skills:
Interpreting complex AI-generated analytics to make informed decisions about learning designs
Continuously refining learning experiences based on AI-generated insights
Strategic Thinking:
Synthesising AI insights with pedagogical theories to create innovative learning strategies aligned to learner needs
Designing adaptive learning ecosystems that leverage AI capabilities
Advanced Critical Analysis Skills:
Assessing AI-generated content for quality, relevance, and alignment with learning objectives and context
Identifying and mitigating potential biases in AI-produced materials
Advanced Prompt Engineering Skills:
Crafting sophisticated prompts that guide AI to produce highly targeted and effective learning content
Iteratively refining prompts based on AI outputs to optimise results
AI-Human Workflow Design:
Designing efficient workflows that optimally blend AI and human efforts
Managing the integration of AI tools into the instructional design process
These elevated skills position the post-AI ID as a strategic learning architect, ethical AI steward, and innovation leader in the educational field.
In the post-AI world, the ID's role becomes less functional and more focused on high-level decision making, creative problem-solving, and ensuring that AI-enhanced learning experiences remain deeply human-centered and pedagogically sound.
Closing Thoughts
An increasing amount of research shows that optimal workplace performance in general (and optimal instructional design performance specifically) happens when we combine the powers of expert human and AI.
While delegating “output tasks” to AI can increases our efficiency (i.e. the speed of the ID process), optimal results which elevate both our efficiency (speed) and effectiveness (impact) emerge when we adopt a Partnership Model of AI-human interaction, where AI serves as an "always-on" thought partner, contributing to functional, creative and strategic decision-making throughout the end to end instructional design workflow.
So the message is clear: adopting a Partnership Model with AI is good news for IDs. But are we, the human instructional designers, ready for this change?
On a practical level, the deep integration of AI in every aspect of our workflow requires us to overhaul our knowledge, skills and tooling; we need to develop a deep understanding of how AI works, how to work with AI (e.g. prompting skills) and how to assess and validate its outputs.
Perhaps more significantly, on a psychological level, the deep integration of AI requires us to relinquish some of our power and control over our workflow and outputs. It also requires a shift from a world where there is a definitive line and clear differentiation between the skills and capabilities of the machine and the skills and capabilities of the human.
Acknowledging that this line is blurred and reconfiguring what this means for our roles, responsibilities and key skills is a huge, existential step for instructional designers and the workforce more broadly.
Whatever the case, two things are clear: AI is here to stay, and IDs who work in partnership with it will lead the field in terms of their efficiency and effectiveness.
Happy experimenting!
Phil 👋
P.S. If you want to get hands-on and experiment with how to integrate AI into your instructional design process, check out my AI Learning Design Bootcamp.