Your Graduates Are Entering
a Labor Market Your Curriculum
Wasn't Designed For.

• Week 1–2: Complete the EDUCAUSE "Teaching with AI" Microcredential Program

  EDUCAUSE's structured 5-module online program (events.educause.edu/teaching-with-ai) is the higher education standard for faculty AI PD. Your AI Lead needs to complete this before they facilitate anything for others. 5–6 hours. Includes sample syllabus language for AI policies.
• Week 1–2: Survey all academic department chairs — one standardized survey

  Questions: Does your department's curriculum include any AI integration? Have your faculty received any AI-related PD in the last 12 months? Does your department have an active industry advisory council? What is your greatest challenge in addressing AI in your curriculum? This is the institutional baseline. Nothing can be measured without it.
• Week 3: Complete MIT OpenCourseWare AI Foundations overview and identify faculty-facing resources

  MIT's free AI course resources (ocw.mit.edu, aiopeneducation.pubpub.org) provide the academic depth your AI Lead needs to speak credibly in faculty conversations across disciplines. Not a substitute for EDUCAUSE, but essential for cross-disciplinary credibility.
• Week 3–4: Meet individually with each college Dean — listen-first agenda

  Not to deliver the agenda. To understand what each Dean believes is already happening, what they're afraid of, and where they see the biggest gaps. This intelligence shapes everything that follows and these relationships determine whether the AI Strategy Lead can actually move the institution.
• Week 5–6: Convene a cross-college Faculty Champions cohort — 1 volunteer per department

  Identify and convene faculty members in each department who are already interested in AI integration. These are not mandatory appointees — they are volunteers. This cohort will be the pilot group for PD, the source of peer learning models, and the internal advocates when skeptical colleagues push back.
• Week 7–8: Deliver AI audit summary to Provost with specific gaps identified and Year 1 action plan

  Not a report on what the university could do. A specific document: here is what is happening now, here is what is missing, here are the 5 highest-priority actions for Year 1, here is the resource requirement, here is the timeline. This document becomes the accountability structure for the next 12 months.

This is the single most important curriculum decision your institution can make. A required, 3-credit course taken by every undergraduate — regardless of major — that covers how AI systems work, how they affect labor markets and human capacities, how they are governed, what ethical obligations users and developers have, and how AI applies to professional practice across industries.

This course should be taught collaboratively — a team-taught model with instructors from computer science, philosophy, social science, and a professional school works well — and reviewed annually. The syllabus should be public.

• How AI systems learn from data — conceptual, not coding-required

  Every graduate should be able to explain to a non-technical audience: AI learns patterns from data, which means it reflects the biases and gaps in that data. This one concept explains most AI failures students will encounter professionally.
• AI and the labor market — honest, data-driven analysis of what is changing

  Use WEF Future of Jobs data, BLS projections, and McKinsey research. Students deserve honest information about the career environment they are preparing for, including which roles are at risk and which are growing.
• Hands-on use of AI tools across professional contexts

  Students should leave having used AI tools in at least 3 different professional scenarios relevant to their intended career. Not just general chatbot use — domain-specific application with critical evaluation.
• Ethics, governance, and accountability of AI systems

  Algorithmic bias, data privacy, AI in criminal justice, surveillance, consent — these are not abstract topics. They are the policy landscape graduates will navigate in every professional role. They need a framework for reasoning about them, not just awareness that they exist.
• How to evaluate AI claims — fact-checking, source verification, uncertainty recognition

  AI hallucinates. AI is confidently wrong. AI can be manipulated via prompt injection. Graduates who know this and have practiced verification protocols are more valuable than those who don't.

Each academic department must integrate a required AI-in-my-field component into its major. This is not the "AI & Society" GenEd — it is discipline-specific, professionally oriented, and tied to current employer expectations in that field. It can be a standalone course, an expanded existing methods course, or a module woven into the senior capstone. The format is less important than the substance.

The AI Strategy Lead coordinates department-by-department audits to determine what exists, what needs to be built, and what the timeline is. Every department should have a plan within 12 months and implementation within 24 months.

A portfolio of short, employer-recognized AI credentials that any enrolled student can earn alongside their major. Partner with EDUCAUSE, Google Career Certificates, AWS Educate, or OpenAI Academy (free training via the National Applied AI Consortium) to minimize development cost. These credentials appear on the transcript and are increasingly recognized by employers as a signal of practical readiness.

Examples: "Data Analysis with AI," "AI-Augmented Research Methods," "AI in Professional Communication," "Responsible AI Deployment in Business." Each completable in one semester, stackable toward a broader certificate.

Courses that teach students to use Microsoft Word, PowerPoint, or Excel at a basic operational level are not preparing college graduates for anything. Students who arrive at university not knowing how to use these tools need direct intervention, not a semester-long course. The credit hours consumed by basic software instruction should be redirected to AI fluency content. Reform these courses entirely into "AI-Augmented Professional Productivity" — building professional workflow skills with current tools at a professional level.

Every research methods course in every discipline — qualitative, quantitative, mixed methods, lab-based — should be updated to include: how AI tools are used in research in that field, what the ethical obligations of AI-assisted research are, how to cite AI assistance, and how to critically evaluate AI-generated literature summaries. This is not a new course — it is an essential update to every existing one. The AI Strategy Lead should provide a cross-disciplinary research methods update guide that departments adapt for their context.

Any career development course using pre-2022 labor market frameworks, outdated career pathway descriptions, or traditional job search strategies without AI context is actively misinforming students. These courses should be rebuilt around: AI's impact on specific career paths, how to leverage AI in the job search, how to position human skills in an AI-augmented marketplace, and how to evaluate AI's risk to your intended career and build resilience.

Start with willing volunteers, not mandates. Identify 1–2 faculty members per department who are already interested. This cohort goes deep in Semester 1, develops AI-integrated curriculum for their courses, documents what works, and becomes the peer learning resource for their departments in Year 2.

• Complete EDUCAUSE "Teaching with AI" Microcredential (5–6 hours)

  events.educause.edu/teaching-with-ai — the higher education standard. Covers AI tools, course design, assignment redesign, AI policy development, and student perspectives. Includes sample syllabus language. Required for all Phase 1 participants.
• Redesign one existing course assignment or assessment using Science of Learning principles

  Not a new course. One assignment, redesigned. Faculty document the design rationale, deliver it, collect student and peer feedback, and present findings to the Faculty Champions cohort. This is the unit of practice that drives behavior change.
• Produce a shareable "AI Integration Guide" for their department

  One-to-two page document: what AI tools are relevant to this discipline, how they recommend integrating them, sample syllabus language for AI transparency policy, one recommended assessment redesign approach. This is the product that makes their work reusable by their colleagues.
• Present at end-of-semester Faculty Champions Showcase

  Publicly recognized. Recorded for the institution's resource library. Celebration of innovation is a cultural signal. These presentations are the most effective PD events your institution can run — peers are more persuasive than administrators.

• All instructional faculty: Required foundation AI PD (minimum 6 hours, verified)

  Not a webinar. A structured, multi-session program facilitated by the AI Strategy Lead and Phase 1 Champions. Faculty who are adjunct or part-time must be included and compensated for participation. Completion tracked and reported in the annual AI Readiness Report.
• Discipline-specific depth tracks for interested faculty (12+ hours, optional stipend)

  Health sciences faculty have different AI integration needs than philosophy faculty. Depth tracks built around disciplinary context produce more usable outcomes than generic AI training. Phase 1 Champions lead these. Internal grant (suggested: $1,500–$3,000 per participant) for curriculum development output.
• Annual recognition: "AI Innovation in Teaching" award — named, celebrated, board-level visibility

  Symbolic recognition that costs almost nothing and signals to the entire faculty community that pedagogical innovation in this area is valued by leadership. Work with Faculty Senate to establish this award category in annual teaching award cycles.

AI Assistance Policy — Tier II

You may use AI tools (such as Claude, ChatGPT, Copilot, or Gemini) on Tier II assignments in this course. If you do, you must submit an AI Contribution Statement alongside your work. This statement must answer: (1) Which tool(s) did you use? (2) What did you ask it to do? (3) What did you change or reject from its output? (4) What portions of the work are entirely your own? Submitting AI-assisted work without this statement is an academic integrity violation. Submitting AI-generated work as if it were entirely your own is an academic integrity violation. The goal is honest, transparent human-AI collaboration — which is exactly what professional environments expect.

The ethical concerns that lead students to decline AI tool use are grounded in documented, real harms: the significant water and energy consumption of large-scale AI inference, labor exploitation in training data annotation, copyright violations in training datasets, and demonstrable job displacement in creative and knowledge-work industries. These are not fringe positions. The students holding them are often among the more analytically rigorous members of a class. Treating their objection as a behavioral problem to be managed is intellectually indefensible and counterproductive.

There is a more important conversation to have with these students — one that most institutions are not yet having clearly: ethical people declining to use a technology does not reduce that technology's power, adoption, or harm. It removes the most thoughtful users from the room where the tool is being deployed and evaluated. The students who understand AI's harms most deeply are precisely the ones whose voices are needed in conversations about how AI is governed, deployed, and constrained. Principled engagement is more powerful than principled abstention.

Implementation

• Student formally declares AI-free status at the start of each semester — in writing, filed with the department

  Not an honor oath requiring surveillance. A formal acknowledgment of the student's chosen track and its requirements. Filed with the AI Strategy Lead and the relevant faculty member.
• All major assignments include a mandatory oral defense component — this is the verification mechanism

  A student who used AI to produce work and committed it to memory can pass a written integrity check. They cannot hold up under a substantive 10–15 minute conversation with a faculty member who knows the subject. The oral defense is both the verification tool and a pedagogically sound practice that strengthens learning regardless of AI use. This should be the standard, not the exception.
• Rubrics explicitly reward demonstrated reasoning and analytical process — not prose polish

  AI-assisted writing consistently produces cleaner, more polished prose than unassisted human writing under deadline. If polish is the implicit grading criterion, AI-Free track students are structurally disadvantaged for a factor that does not reflect their understanding of the material. Reorient rubrics toward the quality of thinking, evidence of engagement with primary sources, and analytical originality.
• Transcript notation: "Demonstrated Unassisted Mastery" on qualifying work

  No extra credit — rewarding the absence of a tool implies that using it is inherently compromised. Instead, students who complete the AI-Free track with successful oral defenses receive a formal notation on their academic record. Graduate programs and research-focused employers will notice. Some will actively value it.
• Students receive explicit, honest instruction on the professional implications of their choice

  Not to coerce a different decision. To ensure the decision is informed. The faculty member or the AI Strategy Lead provides: here is the current wage premium attached to verified AI fluency (56% by current PwC data), here is how hiring in your intended field is currently changing, here is what this choice costs you competitively and what it does not. Students who understand the tradeoff and make the choice anyway are exercising genuine agency. That is worth respecting. Students who make it without the information are not.

• Annual AI labor market training for all career counselors — mandatory, assessed, documented

  A career counselor who last updated their labor market knowledge in 2021 is giving students career advice based on a world that no longer exists. Annual mandatory training using WEF Future of Jobs, BLS Occupational Outlook, McKinsey's State of AI, and PwC's AI Jobs Barometer. Test comprehension. Document completion. Report in the annual AI Readiness Report.
• Establish AI Fluency as a formal graduation milestone — on the transcript

  By graduation, every student should have demonstrated: they can use AI tools relevant to their field, they can critically evaluate AI outputs, and they understand the ethical obligations of AI use in their profession. This is documented in a portfolio or competency assessment. It appears on the academic record. Employers will ask about it. Students will take it seriously once it's credentialed.
• Build college-specific employer advisory councils — meeting at least twice yearly with curriculum input agenda

  The employers hiring your graduates should be reviewing your curriculum — not just speaking at career fairs. Advisory councils for each college (Business, Education, Health Sciences, etc.) that provide formal curriculum feedback twice yearly. Their input should visibly change what you teach within 12 months of their feedback. Otherwise they stop showing up.
• Create an Alumni AI Career Advisors network — formal, structured, connected to students

  Your graduates who are navigating AI disruption in their industries are your most credible career advisors for current students. A named, structured alumni network — quarterly virtual panels, LinkedIn group, one-on-one mentorship matching — provides mentorship that career staff simply cannot replicate. Costs: staff time to coordinate and a Zoom account.
• Audit and rebuild all career counseling materials using 2025–2026 labor market data

  Every career pathway guide, interest inventory, and occupational description should be reviewed against current automation risk profiles. Any material that describes administrative, customer service, data entry, or basic financial roles without addressing AI's impact should be revised. The ACIL provides the labor market intelligence; career services provides the counseling context.

• FERPA-compliant data processing agreements for every AI tool that handles student work

  When students submit assignments through AI platforms, those are education records under FERPA. Written DPAs are required — not assumed. "We use it informally" is not a legal defense.
• Prohibited data types defined, documented, and communicated to all faculty

  IEP information, accommodation plans, health records, behavioral records, disability documentation, and personally identifiable student information may NEVER be uploaded to an unapproved AI tool. State this explicitly. Faculty need clarity, not inference.
• Vendor contracts prohibit training AI models on student work

  Many free AI tools improve their systems using user-submitted content. Student essays, code, and research are not training data for a private company. Vendor contracts must explicitly prohibit this use. If a vendor will not sign that clause, the institution should not adopt the tool for classroom use.
• AI-approved tools list maintained by the AI Strategy Lead and published to faculty

  Faculty should not be independently selecting tools for course use without vetting. An approved list provides the framework faculty need and the protection the institution needs. Updated annually.
• Incident response protocol for accidental uploads of protected information

  A faculty member will upload a document containing accommodation information or a student's diagnosis. The protocol should exist before it happens: who is notified, what is asked of the vendor, how the student and parents are informed, what is documented.
• Annual vendor compliance audit

  Vendor data practices change. What was acceptable in 2025 may not be acceptable in 2027. An annual review of each approved tool's current data-use practices, terms of service, and privacy policies is required — not assumed.

• AI Strategy Lead: Named, active, adequately resourced (at least 50% appointment equivalent)

  Y/N and the person's name. If N, this is the root cause of every other failure on this list.
• % of departments with completed AI curriculum audit (target: 30% Y1, 80% Y2, 100% Y3)

  Named departments. Not approximations. If Engineering is at 100% and Education is at 0%, that gap is the next conversation.
• "AI & Society" GenEd requirement: In development (Y1), launching (Y2), required for all majors (Y3)

  One of the highest-impact single decisions your institution can make. If this is not moving, name the barrier specifically.
• % of instructional faculty who completed meaningful AI PD (target: 20% Y1, 60% Y2, 90% Y3)

  Track by college. If one college is at 80% and another is at 10%, the gap needs targeted support — not another all-faculty memo.
• AI Fluency graduation milestone: Drafted (Y1), piloted (Y2), institutionalized (Y3)

  When employers ask your Career Center "do your graduates have AI fluency?" — what do you say? This milestone is the answer.
• # of students who completed AI micro-credentials this year, tracked by major and demographic

  If only STEM students are earning AI credentials at a university, you have an equity problem and a liberal arts employability problem simultaneously.
• Employer Advisory Councils: Active in each college, met at least twice, curriculum feedback documented and actioned

  If the council exists but has not changed anything in your curriculum, it is a ceremonial function. Ceremonial functions don't justify the time investment.
• Career center staff: Completed annual AI labor market training, all counseling materials updated

  Ask a career counselor: "What is the automation risk profile for a student who wants to be a paralegal? An accountant? A journalist? A nurse?" Their answers tell you whether the training landed.
• AI Readiness Report: Published publicly, reviewed by Board of Trustees, includes year-over-year comparison

  If this report does not exist, you have not created public accountability for your commitments. Peer institutions that publish theirs will use it as a competitive advantage.
• Accreditation documentation: AI integration efforts formally documented and included in accreditation records

  This is also your risk management strategy. Don't wait for accreditors to ask — document proactively and continuously.
• K-12 partnership: At least one active partnership with local school district — named, active, mutual benefit

  The university that positions itself as the community's resource for AI literacy builds goodwill, enrollment pipelines, and community trust simultaneously. The partnership doesn't need to be large. It needs to exist.