Artificial intelligence has become one of the most talked-about topics in education, and for good reason. New AI tools can generate text, summarize research, create images, write code, brainstorm ideas, and even help draft assessments in a matter of seconds. For instructional designers, those capabilities can feel both exciting and overwhelming because the technology seems to be changing faster than most institutions can create policies around it. Some people see AI as a productivity booster, while others worry that it may eventually replace parts of the design process.
In practice, the most useful role for AI is much less dramatic. AI works best as an assistant rather than as an instructional designer because it can accelerate repetitive tasks, spark ideas, and reduce the time spent staring at a blank screen. It cannot replace instructional judgment, learner analysis, faculty collaboration, accessibility review, or the human relationships that shape successful learning experiences. Those are the parts of instructional design that require context, experience, and professional decision-making.
The more useful question, then, is not whether instructional designers should use AI. It is how we can use it thoughtfully, responsibly, and with enough skepticism to know when it is helpful and when it is simply producing confident nonsense. When used well, AI becomes another tool in the instructional designer’s toolbox, one that can free up time for deeper design work. This article looks at practical ways to use AI as an instructional design assistant while keeping human judgment firmly at the center of the process.
AI Is a Tool, Not a Replacement
Whenever a new technology emerges, predictions tend to swing between extremes. Some people assume it will change everything overnight, while others believe it will eliminate entire professions. The conversation around generative AI has followed the same pattern, with plenty of headlines suggesting that instructional designers may soon be replaced by automated systems. That prediction overlooks how much of instructional design depends on human judgment and collaboration.
Instructional design is not simply content generation. It involves understanding learners, working with subject matter experts, aligning objectives and assessments, supporting accessibility, selecting appropriate technologies, and making decisions within institutional and disciplinary contexts. AI can assist with pieces of that work, but it does not independently understand why one instructional choice is better than another in a particular situation. It can offer suggestions, but the designer still has to determine whether those suggestions make instructional sense.
Thinking of AI as an assistant creates a much healthier frame. A calculator did not replace mathematicians, and presentation software did not replace teachers. In the same way, AI is more likely to change how instructional designers complete some tasks than to eliminate the need for instructional design itself. The real shift is in workflow, not in the value of the profession.
Start With a Blank Page Less Often
One of AI’s greatest strengths is helping with the blank-page problem. Most instructional designers have opened a document and wondered where to begin, whether they were drafting learning objectives, writing module introductions, developing discussion prompts, or organizing a new course structure. The first draft often requires more mental energy than the later revisions. AI can make that first step much easier.
For example, an instructional designer might ask AI to generate several possible module outlines, draft a welcome announcement, suggest scenario ideas, or propose different ways to organize a complex topic. The result should not be treated as finished work. Instead, it provides material to react to, revise, combine, or reject. That is often enough to get the design process moving.
This is where AI can support creativity rather than replace it. A generated draft may include ideas the designer would never use, but it may also introduce a structure or example that sparks a better solution. Sometimes the value is not in what AI writes. It is in what the designer thinks of next.
Brainstorm Before You Build
Instructional design involves constant problem-solving, and most design questions have more than one possible answer. How should a difficult concept be taught? What kind of activity would help learners apply it? How could an assessment become more authentic without becoming unmanageable? AI can be useful here because it can generate multiple possibilities quickly.
Instead of asking AI for one “best” activity, instructional designers can ask for several different approaches and compare them. One idea might emphasize collaboration, another might use a scenario, and a third might suggest a reflective activity. None may be perfect, but together they can broaden the range of options under consideration. That kind of brainstorming can be especially useful when the designer feels stuck or has been working with the same types of activities for too long.
The important part is that AI remains a source of possibilities rather than the final decision-maker. The instructional designer still considers learner needs, course constraints, accessibility, workload, and alignment. AI can widen the menu. The designer still chooses what belongs on the plate.
Use AI for Routine Work
Instructional designers spend a surprising amount of time on tasks that are necessary but not especially creative. Drafting announcements, reformatting content, summarizing notes, creating checklists, revising repetitive wording, and organizing meeting materials can consume hours that might otherwise be spent on higher-level design work. AI is particularly well suited to assisting with these kinds of tasks. It can often produce a usable starting point in seconds.
That time savings matters because efficiency should create room for better design, not simply more work. If AI helps reduce the time spent on routine writing, instructional designers can spend more time reviewing accessibility, improving assessments, consulting with faculty, or analyzing learner needs. Those are areas where human attention adds much greater value. The goal should be to shift effort toward the parts of the job that most directly affect learning quality.
There is also a practical benefit for teams with heavy workloads. Instructional designers often manage multiple course projects at once, each with its own deadlines, faculty needs, and institutional requirements. Using AI strategically can reduce some of the smaller administrative burdens that accumulate across those projects. Even modest time savings can add up quickly.
Improve Existing Content
AI can be especially useful when the content already exists but needs improvement. Instructional designers frequently receive materials that are overly dense, repetitive, jargon-heavy, or written for experts rather than learners. Asking AI to suggest clearer wording, reorganize sections, or generate alternative explanations can speed up the revision process. It can also help identify places where the same idea is repeated unnecessarily.
That does not mean AI should be trusted to rewrite content without review. A passage can sound polished while still being instructionally weak or factually inaccurate. The designer must still evaluate whether the revision preserves meaning, uses appropriate terminology, and fits the audience. In some disciplines, even a small wording change can alter technical accuracy.
The most productive mindset is to treat AI like an editor making suggestions. It can point toward simpler phrasing or a clearer structure, but the instructional designer decides whether those changes actually improve the learning experience. That distinction keeps the human in control of both meaning and quality.
Ask for Options, Not Final Answers
One of the best ways to use AI is to ask for multiple possibilities rather than a single finished product. If you ask for one discussion prompt, you may get something usable, but you may also get something generic. If you ask for five prompts that emphasize analysis, application, and real-world decision-making, you have more material to compare and refine. The same principle applies to case studies, assessment ideas, examples, and module structures.
Generating options encourages instructional designers to stay engaged in the design process. Instead of accepting the first response, they can combine ideas, identify patterns, and decide which approach best fits the learning objective. This makes AI part of a conversation rather than a vending machine for course content. It also reduces the temptation to treat a polished answer as automatically correct.
Sometimes the best result is not any of the suggestions AI generates. The real benefit may be that one suggestion sparks a completely different idea. In that sense, AI can function much like a brainstorming partner who is always available and occasionally says something useful enough to get the room thinking.
Human Judgment Still Matters Most
There are many areas where instructional designers should remain firmly in control. Learning objectives must align with assessments. Activities must support learner needs. Accessibility decisions require context. Feedback should reflect the purpose of the assignment, and technology choices must fit institutional policies and learner capabilities. These are not decisions that should be handed over to AI simply because the tool can produce an answer.
AI can recommend, but it does not carry professional responsibility. If an assessment is misaligned, a resource is inaccessible, or a policy is misrepresented, the instructional designer is still accountable for the final product. That makes human review more than a nice extra. It is a required part of responsible AI use.
This is also where professional experience becomes more important, not less. The better an instructional designer understands learning theory, accessibility, assessment, and course design, the easier it is to recognize when an AI-generated suggestion is weak. AI may increase productivity, but expertise is what determines whether that productivity leads to better learning.
Accuracy Must Be Verified
Generative AI can produce incorrect information while sounding completely confident. That combination is especially dangerous in instructional design because polished language can make errors harder to notice. References may be fabricated, definitions may be slightly wrong, and examples may contain details that do not hold up under scrutiny. Any AI-generated content intended for learners should therefore be reviewed carefully.
Verification becomes even more important in high-stakes fields such as healthcare, law, science, engineering, or public policy. In those areas, inaccurate information may have consequences far beyond a quiz score. Instructional designers should verify facts, citations, terminology, accessibility guidance, and institutional policies before anything reaches learners. A fast draft is useful only if it becomes an accurate final product.
This is one of the reasons AI should be treated as an assistant rather than an authority. It can help generate material quickly, but speed is not the same as reliability. The designer still has to do the thinking that determines whether the content deserves to stay.
Privacy and Institutional Policies Matter
AI use also raises questions about privacy, data security, and institutional policy. Instructional designers may work with student information, unpublished course materials, internal documents, assessment data, and other content that should not automatically be entered into public AI systems. Knowing what information can safely be shared is part of responsible use. Institutional guidance should always take precedence over convenience.
This is particularly important when working with student work. Personally identifiable information, protected educational records, confidential faculty materials, and sensitive research data should not be copied into AI tools unless the institution has explicitly approved that use. The fact that a tool makes a task easier does not mean it is appropriate for every type of information. Convenience does not override privacy.
Responsible AI use therefore begins before the prompt is written. Instructional designers need to know which tools are approved, what data they can process, and what restrictions apply. Those decisions are part of professional practice just as much as assessment alignment or accessibility review.
AI Can Support Accessibility, but It Cannot Replace Review
AI can also assist with accessibility work. It can generate draft alt text, suggest plain-language revisions, help edit captions, create preliminary accessibility checklists, and offer alternative descriptions of complex visuals. These uses can reduce the time required for repetitive accessibility tasks and help instructional designers identify issues earlier in the design process.
The key word, however, is draft. Accessibility depends heavily on context, and AI does not always understand the instructional purpose of a resource. A technically accurate image description may still fail to communicate what learners actually need to know from the image. A caption may contain all the words but still mishandle technical terminology or punctuation.
Human review remains essential because accessible design is about learner access, not simply compliance with a checklist. AI can speed up parts of the process, but the instructional designer still has to evaluate whether the result truly supports the learner. That is another example of where automation helps most when paired with professional judgment.
The Human Skills Become More Important
As AI handles more routine tasks, the human skills that instructional designers bring to the profession become even more visible. Building trust with faculty, asking the right questions, understanding learner needs, facilitating difficult conversations, and balancing competing priorities are not easily automated. These skills often determine whether a course development project succeeds. They are also the parts of the work that faculty and learners remember.
Instructional designers frequently serve as translators between subject matter expertise, institutional requirements, technology, and learner needs. That role requires empathy, communication, and an ability to read situations that AI does not possess. A tool can generate a course outline, but it cannot manage the conversation when a faculty member is worried about changing an assignment they have used for ten years. It can suggest options, but it cannot build the relationship needed to make change possible.
In that sense, AI may push the profession toward the work instructional designers should have been prioritizing all along. If technology takes on more routine drafting and formatting, designers can spend more time on consultation, problem-solving, strategy, and learner-centered decision-making. Those are the parts of instructional design that are hardest to automate and most important to preserve.
Final Thoughts
Artificial intelligence is changing instructional design, but the most useful change is not replacing instructional designers. It is giving them another tool for drafting, brainstorming, revising, organizing, and handling routine tasks more efficiently. Used thoughtfully, AI can reduce friction in the design process and create more time for work that requires human attention.
The strongest instructional designers will not be the ones who avoid AI completely, nor will they be the ones who hand every task over to it. They will be the ones who understand where AI adds value, where it introduces risk, and where professional judgment must remain in control. That balance will matter more as the technology continues to evolve.
AI can help generate the first draft, suggest alternatives, or shorten the time required for routine work. It can be a useful assistant and, occasionally, a surprisingly good brainstorming partner. But the instructional designer still decides what supports learning, what meets standards, what respects learners, and what belongs in the final course.
That is the part of the job AI should assist, not replace.
References
Clark, R. C., & Mayer, R. E. (2023). e-Learning and the science of instruction: Proven guidelines for consumers and designers of multimedia learning (5th ed.). Wiley.
Mollick, E. (2024). Co-intelligence: Living and working with AI. Portfolio.
Nilson, L. B., & Goodson, L. A. (2021). Online teaching at its best: Merging instructional design with teaching and learning research (2nd ed.). Jossey-Bass.
UNESCO. (2023). Guidance for generative AI in education and research. UNESCO.
Wiggins, G., & McTighe, J. (2005). Understanding by design (Expanded 2nd ed.). ASCD.