Example · Instructional Designer
68% of your week as an instructional designer is still yours.
AI now builds course content from a syllabus. Your standing with faculty stays with you.
Make my planA typical week at the middle pace. Your plan uses your own tasks.
Is instructional design dying because of AI?
In a typical week at the middle pace, 30% of this work needs your judgment and 38% is work you do using AI. AI could do about 32%. Production is 32% of your week: the LMS shell, the module text, the video edit, the captions, the first draft of the quiz. A model now makes all of it from a syllabus and a slide deck.
Read more
LMS vendors sell this as a feature, and faculty may prefer the machine's version of their own course because it sounds like them. Under the moderate scenario, an online division of modest size stops paying four people to do that within three years, mostly by not refilling the next seat that empties.
What makes you hard to replace is the meeting where a professor who has taught the same lecture for twenty years decides, with you in the room, to change it. Federal rules require a named instructor in regular, substantive contact with students, and a federal accessibility deadline now sits on public universities with a date attached and the Justice Department's own finding that AI cannot do the remediation. Nobody can build a course for a faculty member who has not agreed to it, and nobody can sign the university's name to a course a screen reader cannot use. The plan moves your week toward those two rooms, and it makes the hard conversations easier by giving you authority you don't have today. Note that the biggest threat to your seat is the enrollment numbers, scored separately below.
The week, task by task
Who does each hour of a typical 40-hour week.
See every task
Needs your judgment.
- Faculty consultation: intake, design meetings, negotiating changes to the course14%
- Program-level curriculum mapping, accreditation and RSI documentation, new-program proposals6%
- Faculty development workshops, in person and virtual5%
- Team, governance and committee meetings5%
AI does the slow parts. You still own it.
- Accessibility remediation: alt text, heading structure, PDF tagging, caption correction, the accessibility-checker queue10%
- Assessment design: quiz items, rubrics, objective alignment8%
- Development pipeline: tracking, SME contracts, launch calendar6%
- Quality review against the division rubric; course-ready sign-off6%
- + 2 smaller tasks8%
AI could do most of it.
- Build LMS course shells from the division template: modules, pages, settings, due dates12%
- Multimedia: lecture-capture editing, graphics, machine captions, interactives10%
- Draft and edit module text, overviews, objectives, instructions from faculty material10%
Why some tasks still need a person
You sit with faculty and turn "put my course online" into a design
This rests on context and relationships. Faculty own the curriculum under shared governance, and the accreditor holds the institution to that. A model given the syllabus builds the course, but it cannot get the professor to agree that his week-9 lecture should be two weeks, or notice that he is frightened of the discussion board. Persuading a tenured faculty member takes standing, and you earn standing in meetings.
You are the reason a course launch is defensible on accessibility
The accountability here is borrowed from the university's duty under ADA Title II, the disability law for public bodies. The 2024 DOJ (Justice Department) rule requires WCAG 2.1 AA, the web accessibility standard, for public entities, public universities included, with a compliance date now of April 26, 2027 for larger entities and April 26, 2028 for smaller ones. DOJ's April 2026 extension states that generative AI "does not yet reliably automate the remediation of inaccessible content at scale." Someone must decide what gets fixed, what gets replaced, and what the professor is told.
You check that a course has a real instructor in it
The accountability here is borrowed from Title IV, the federal student aid law. 34 CFR 600.2 defines distance education as requiring "regular and substantive interaction" with an instructor "responsible for delivering course content." A course without that is correspondence education and loses federal aid eligibility. The check is a per-course design judgment a program director wants a human to have made.
You fix the assessment when faculty say students are using ChatGPT
This depends on context. Only 24% of faculty have redesigned assessments around AI (Tyton Partners, 2026), and grading is where faculty rate AI least effective (Anthropic's education report). The redesign needs the discipline, the professor's tolerance for change, the accreditation constraints, and the trust to say "your essay prompt is the problem."
You catch the wrong quiz items and the wrong textbook edition in AI drafts
What protects this is verification, and nobody else in the chain will do it. The professor may like the draft, but the model cannot know which edition the bookstore stocks. When drafting costs little, the person who knows what to check becomes more valuable.
Good to know
What is hard about this field, and what you can do about it.
The hard truth
The experts you serve can now build their own materials.
Faculty and subject experts used to need you to turn their knowledge into a course. More of them now draft materials with AI.
Sources (1)
- Claude.ai conversations to develop instructional materials rose from 0.2% to 1.5% within about eight months. Anthropic Economic Index, 2025-09
What can an instructional designer move into?
Plan A is the best move inside the work you do now.
Grow in the job you have
The main focus. Most weekly sessions build toward it.
- Each week
- 4 sessions
- Ready in
- 4 to 6 months for the program
Why it fits
Why you. You likely already do the remediation and already have the hard conversation, and often nobody else in the division owns it. The date moved once and DOJ plans further rulemaking, so the institution needs someone who tracks the rule and does more than tag PDFs. DOJ's reasoning for the extension matches your job description. It says staff lack technical expertise, and AI "does not yet reliably automate the remediation of inaccessible content at scale."
The gap. You have no inventory by conformance, no triage rule, no signed escalation path and no accessibility credential yet.
The trade-off. This is the hardest part of the work, and the plan asks for more of it. Borrowed accountability is the weakest kind. The university can hand remediation to a vendor and reporting to IT, and your program becomes theirs. If DOJ softens the standard, the duty stays but the urgency goes. Your name on the report also means your name on the finding when the number is bad.
2 other moves that fit
- Be the assessment-redesign consultant faculty ask forReady in about 3 months
- Lead one program from proposal to launchReady in about 12 to 18 months
What's a plan B for instructional designers?
Plan B is meant to be different from this job. It was picked because skills from this role carry over.
Get ready for a new kind of job
A second option. A newer job AI is creating that these skills already fit.
Aligning assessments to outcomes with faculty is exactly what AI tutoring products need to teach accurately and fairly.
- Each week
- 1 session
- Outlook
- Emerging
- Pay
- $70.94 to $84.42 an hour (temp contract)One Aquent temp posting, Oct 2025; freelance AI tutor training pays $6 to $65 an hour.
2 other options
- Conversation designerYou already sequence content and feedback for learners, which maps onto designing the turns and handoffs of an AI conversation.
- AI enablement and adoption leadFaculty already come to you when tools change. Adoption work lasts when it is tied to their teaching, such as assessment redesign.
In your own plan you can switch to any of the 47 roles.
The 13 weeks
Small steps that add up. One plan for both: most sessions build Plan A, one a week builds Plan B.
By week 13 you will have
AFor Plan A
- At least three assessment consultation cases
- A memo asking for a new title
- A published AI tutor case study
BAnd for Plan B
- A draft tutor evaluation case study
- A published AI tutor case study and one application
- 1Measure your week
- 2Build the accessibility inventory
- 3Set a triage rule and log
- 4Check courses, plan consultations
You finish withMore of the QM course done, with your notes
- 5Make your first log entry
- 6Run your first consultation
- 7Propose the program
- 8Gather early evidence
You finish withA meeting date and a list of active program proposals
- 9Fix courses and follow up
- 10Build the dashboard
- 11Count the term
- 12Make your ask
- 13Decide your next step
You finish withA one-page review of your 13 weeks
Three courses from the plan
- ADA Title II web rule fact sheet and April 2026 Interim Final Rule (91 FR 20902)US Department of Justice, ada.gov / federalregister.gov · Free
Yes, Week 1.
- HHS Section 504 final rule (web and mobile accessibility)HHS, federalregister.gov · Free
Ask your ADA coordinator whether HHS dates moved in step with DOJ's.
- 34 CFR 600.2, definition of distance education (regular and substantive interaction)eCFR · Free
Yes, Week 2.
Checked 2026-09-22.
The fuller picture
Risks beyond AI: highEnrollment risk
This one has nothing to do with AI and runs on a shorter clock. The high-school graduate peak has passed, graduate enrollment is flat, and new federal loan caps change what graduate students can borrow for the programs online divisions sell. A cut driven by enrollment arithmetic is one no design skill prevents.
Where this could be wrong
- The shares are guesses until Week 1. If your log shows far more build time and far less faculty consultation, junior openings shrink faster than we show, and the accessibility move becomes an application rather than an ask.
- We assumed a public institution. At a private university Title II does not apply; Section 504 does, through federal funding. The duty remains, but the calendar pressure softens.
- We assumed a consultative design model. Some divisions run a course factory: designers build master courses with a paid SME (subject expert) and adjuncts teach them. There, faculty consultation is transactional and the work that holds up shrinks to program paperwork. If that is your division, our numbers are too optimistic by 8–10 points.
- The federal rule may change substance, not just dates. DOJ said in April 2026 that it plans further rulemaking. The accessibility move rests on a date and a number, and both could move again.
Also matched: instructional designer, instructional design, learning designer. Research checked 2026-09-22. Page updated .