How Instructional Designers Are Using AI (and Where It Falls Short)

AI is changing instructional design quickly.

In a 2025 study of 232 instructional designers by a national talent development association, nearly two-thirds of those who used AI had started using it within the previous year.

Among AI users...

37% said it greatly reduced the time they spent designing courses
70% said it improved the quality of their work.

Still, about one in five designers in the study didn't use AI, mostly because they didn't trust its results. Both perspectives are understandable. AI can save time, but it can also make mistakes.

Here's how instructional designers use AI, where it can fall short, and how to review its work before learners see it.

Seven Practical Ways Instructional Designers Use AI

Designers often use AI to summarize information, draft content and brainstorm ideas. Here are seven common examples, along with what to check in each one.

Task	How AI can help	What to check
Course outlines	Creates a first-draft structure from source material	Does the outline fit your learners and the goal of the course, or is it too generic?
Learning objectives	Suggests measurable learning objectives	Do the objectives match the actual skills learners need on the job?
Quiz and assessment drafts	Drafts questions and answer feedback	Are the answers correct? Are wrong options believable, and do the questions test useful skills rather than simple recall?
Scenarios and dialogue	Suggests realistic situations and branching choices	Would the situation happen in real life? A subject matter expert may spot details that don't ring true.
Summarizing expert content	Condenses long documents and interview notes	Has it missed important context or added details? Is the content safe to share with the tool?
Alt text and captions	Drafts image descriptions and transcripts	Does the description explain the image's purpose? Are captions accurate, especially for technical terms?
Voiceover and translation	Creates narration and translated versions quickly	Are pronunciation, tone and terminology right for your audience?

An industry survey of 421 learning and development professionals, conducted by an AI video company, found that the most common uses were voice generation (63%), quiz and content drafting (60%) and video (52%).

Only 19% used AI to evaluate training.

That difference is important: AI can help teams create learning materials faster, but people still play a major role in checking whether training works.

AI Features Are Now Part of Authoring Tools

You don't always need a separate AI chatbot. Popular course-authoring tools now include AI features.

As of October 2026, Articulate 360 offers features such as drafting courses from prompts or source material, checking accessibility, drafting alt text, creating captions and transcripts, and generating text-to-speech narration.

Some advanced features, including translation and AI avatar video, use credits that cost extra.

These features can speed up the first draft, but they don't replace a final review.

Independent reviewers note that AI-generated alt text and navigation text still need human attention. Complex course interactions also still require substantial hands-on work.

Where AI Can Fall Short

Accuracy: AI can present incorrect information as if it were true. In technical, medical, legal or compliance training, mistakes can have serious consequences. Ask a subject matter expert to verify the facts and procedures.
Assessment quality: AI can draft quiz questions, but each question still needs to measure a learning objective. Check for incorrect answer keys, obviously wrong choices and questions that test memory instead of real-world performance.
Accessibility: Automated checks can find some issues but miss others. In April 2026, the U.S. Department of Justice extended its ADA Title II web accessibility deadlines and cited the limits of generative AI for fixing accessibility problems as one reason. Treat AI accessibility features as a first step, not a complete review.
Privacy and confidentiality: Sharing employee data, customer information or unreleased company material with a public AI tool may violate laws or workplace policies. Check which tools your organization has approved.
Copyright and licensing: AI-generated images, voices and text can raise questions about ownership and permitted use. Understand your organization's rules before publishing them.
Bias: AI may repeat stereotypes in names, images and scenarios. Review who is represented and how people are portrayed.
Generic content: AI often produces predictable material. It can't interview your learners or fully understand the performance problem at your organization. That analysis is still your responsibility.

A Human Review Checklist for AI-Assisted Content

Before learners see content created with AI, use this checklist to catch errors and make sure the material is ready.

Check the facts: A subject matter expert has reviewed claims, numbers and procedures.
Check alignment: Each activity and quiz question connects to a learning objective, and feedback explains why an answer is right or wrong.
Check answer keys: Correct answers are accurate, and incorrect options are plausible but still wrong.
Check accessibility: Test with a keyboard and screen reader, not just an automated checker. Make sure alt text explains the image's purpose and captions are accurate.
Protect data: Don't put personal or confidential information into a tool your organization hasn't approved.
Confirm usage rights: Make sure images, voices and text can legally be used for your project.
Review representation: Check that names, images and scenarios are fair and inclusive.
Review the tone: Make sure the content sounds like your organization and fits your learners.
Test with learners: Ask a few real learners to try the content, then fix anything that confuses them.

What AI Means for Your Instructional Design Career

If AI can draft a course outline in seconds, what does that leave for instructional designers? The work that requires judgment, context and a clear understanding of people.

Identify the real problem. Is training the answer, or would a job aid, a process change or better tools help more?
Understand the learners. Find out what people already know, what makes the work difficult and what will help them apply new skills.
Connect goals, practice and assessment. Make sure course objectives, learning activities and assessments work together and reflect what people need to do on the job.
Evaluate quality. Spot content that is inaccurate, inaccessible or ineffective, including AI-generated drafts.
Measure results. Find out whether training improves performance. This remains one of the less common uses of AI in learning and development.

Instructional designers get the most from AI when they understand the fundamentals. Knowing what effective learning looks like helps you use AI to work faster without sacrificing quality.

Build the Skills AI Can't Replace

Oregon State's E-Learning Instructional Design and Development Certificate helps you build these core skills, including learner and needs analysis, measurable learning objectives, assessment design, accessibility and evaluation. You'll receive feedback from experienced instructors as you apply what you learn.

The program is 100% online, takes about 3 to 6 hours per week, and has no application requirement.

Want more practical advice?

Frequently Asked Questions

Will AI replace instructional designers?

AI is changing how instructional designers work, especially when drafting and producing course materials. But people are still needed to analyze learning needs, design for real-world performance, check accuracy and accessibility, and measure results.

What are the best AI tools for instructional design?

Many designers use general AI assistants to brainstorm and draft content, along with AI features built into course-authoring tools for tasks such as captions, alt text and narration. Choose tools that your organization has approved for the type of data you'll use.

Can AI write alt text and captions?

Yes, AI can create a first draft and save time. However, alt text may describe an image without explaining its purpose, and captions may get technical terms wrong. Review both before publishing.

Is it ethical to use AI to create course content?

It can be, as long as you use it responsibly. Check facts, protect private information, confirm that you have the right to use generated content, look for bias, and be transparent with stakeholders about how AI was used.