Palmer, K., 2026. Data Shows AI ‘Disconnect’ in Higher Ed Workforce [WWW Document]. Inside Higher Ed. URL https://www.insidehighered.com/news/workplace/staff-issues/2026/01/13/data-shows-ai-disconnect-higher-ed-workforce (accessed 1.18.26).

Robert, J., 2026. The Impact of AI on Work in Higher Education [WWW Document]. EDUCAUSE. URL https://www.educause.edu/research/2026/the-impact-of-ai-on-work-in-higher-education (accessed 1.18.26).

Teaching and learning. Main takeaways from the Educuase report

1. Faculty and staff already use AI for teaching-related work

• Expect routine use, not experimentation. 94% used AI tools for work in the past six months.
• Expect frequent use. Among recent users, 73% use AI daily or weekly.
• Expect broad task coverage. 54% used AI for eight or more task types in six months.

2. Teaching work concentrates on “content production” tasks
   High-frequency teaching and learning tasks where people already rely on AI:

• Brainstorming lesson ideas and examples (63%)
• Drafting emails to students and colleagues (62%)
• Summarizing long readings, meetings, or guidance docs (61%)
• Proofreading and copyediting (56%)
• Creating presentations (47%)
• Creating training materials (38%)
• Creating to-do lists and work plans (36%)
• Writing procedure guides such as SOPs (35%)
• Writing proposals (34%)

Implication for your courses

• You should treat these as baseline workflows. Design guidance and templates for these tasks first because adoption already exists.

3. Faculty show a clear teaching-specific pattern: more AI use for learning activities and assessment

• 63% of faculty used AI to create learning activities or assessments, compared with 32% of staff.

Implication for teaching quality

• You need a shared standard for assessment design with AI, including item validity checks, alignment to learning outcomes, and safeguards against bias and superficial items.

4. Policy and guidance do not match the pace of classroom-facing use

• Only 54% know the policies or guidelines for work-related AI use, even though use is widespread.

Direct teaching risk

• Inconsistent course expectations across instructors, inconsistent student guidance, and uneven handling of data, accessibility, and citation.

What to do in teaching and learning

• Publish a one-page “AI in teaching work” guide for instructors, then embed it in the LMS course shell template.

• Require each course syllabus to state what counts as acceptable AI use for student work.

5. Institutions lean permissive or neutral, but clarity stays low

• Where policies exist, many describe them as permissive (47%) or neutral (30%).

• Only about half feel confident that current policies give enough clarity.

Implication

• Permissive policies without concrete examples lead to uneven practice. You need course-level examples and rubrics, not policy language alone.

6. The opportunity set targets workload reduction and better use of data
   Most promising opportunities, relevant to teaching and learning:

• Automating repetitive processes (70%)
• Offloading administrative burdens (65%)
• Analyzing large datasets, including qualitative data (60%)
• Generating insights for data-informed decisions (53%)
• Real-time analytics and visualization (51%)
• Creating learning tools (45%)
• Aiding course design and development (44%)
• Accelerating research progress (41%)

What to prioritize first in teaching and learning

• Feedback workflow: draft feedback, then require instructor review plus a quality checklist.
• Course design workflow: outcomes, weekly plan, activity bank, assessment blueprint.
• Accessibility workflow: alternative text, plain-language rewrites, captioning support, readability checks.

7. The top risks map directly to teaching quality and academic standards
   Most urgent risks with direct teaching impact:

• Increased misinformation (55%)
• Use of data without consent (52%)
• Loss of fundamental skills requiring independent thought (51%)
• Insufficient data protection (51%)
• Copyright and intellectual property violations (47%)
• Student AI use outpacing faculty and staff AI skills (47%)
• Surface-level work proliferation (46%)
• Privacy and security law or policy violations (45%)
• Inability to evaluate AI-generated content (41%)

What to do in teaching and learning

• Put a verification routine into every AI-supported teaching workflow: source check, evidence check, bias check, accessibility check.
• Create a “no sensitive data” rule for student information and assessment materials unless you use institution-approved tools.

8. The main blockers: pace, expertise, and best practices
   Top challenges:

• AI’s pace of change (60%)
• Lack of AI expertise (55%)
• Lack of best practices (48%)
• Lack of time to learn new skills (46%)

What works better than general training

• Role-based micro-training for instructors, 3 modules of 20 minutes each:

  • Designing activities and assessments with AI, with validation checks

  • Feedback and grading support, with rubric alignment and bias checks

  • Student guidance and integrity, with transparent disclosure and documentation

9. Shadow tool use creates teaching and learning exposure

• 56% used tools not provided by the institution for work tasks.
• 23% of faculty said their institution provides no access to any AI tools they want to use.

Why this matters for teaching

• Instructors will default to personal accounts for course materials, drafts, and student communications unless the institution provides vetted tools.

Action that reduces risk fast

• Provide at least one approved assistant for writing and summarizing, plus an approved tool for slide creation.

• Publish a short list of approved tools and a short list of prohibited uses, tied to data types and course artifacts.

If you want, share one course type you care about (intro lecture, methods, seminar, large enrollment, online). I will translate these takeaways into a course-ready package: syllabus language, assignment templates, a student AI use policy, and a rubric add-on for AI-assisted work.

Main takeaways from the Inside Higher Ed article “Data Shows AI ‘Disconnect’ in Higher Ed Workforce” (Kathryn Palmer, January 13, 2026) (insidehighered.com)

• Widespread AI use, weak policy awareness

  • 94% of higher ed employees report using AI tools for work.

  • Only 54% say they know their institution’s AI policies and guidelines.

  • Only about half feel confident using AI for work even when policies exist. (insidehighered.com)

• The “disconnect” creates institutional risk

  • The article links low policy awareness and low confidence to risks in data privacy, security, and data governance. (insidehighered.com)

• Most campuses say they have an AI strategy, but practice varies by role

  • 92% report a work-related AI strategy, often including pilots, risk and opportunity evaluation, and encouragement to use AI.

  • Many people in roles most likely to shape policy still report low awareness of policies, including executive leaders, managers, IT, and privacy and security staff. (insidehighered.com)

• Shadow AI use is common

  • 56% report using AI tools not provided by their institution for work tasks. (insidehighered.com)

• AI use stays mostly voluntary, yet intention to keep using AI stays high

  • 89% say they are not required to use AI tools for work.

  • 86% say they want to use, or keep using, AI tools in the future. (insidehighered.com)

• Employee sentiment sits between interest and concern

  • 81% show at least some enthusiasm, split between enthusiastic and mixed caution plus enthusiasm.

  • 17% report caution.

  • Perceptions of leaders mirror this pattern, with many seen as enthusiastic or mixed. (insidehighered.com)

• A key critique: the sample and answer options may inflate “enthusiasm”

  • A quoted expert notes only 12% of respondents were faculty, so results may reflect staff and administrative use cases more than teaching-focused ones.

  • The expert also argues the survey’s response options may push respondents toward “mix of caution and enthusiasm,” which can blur what people feel. (insidehighered.com)

• Risks and opportunities both register as broad and high

  • 67% select six or more urgent risks, including misinformation, data used without consent, skill loss tied to independent thought, students outpacing staff skills, and job loss.

  • 67% select five or more promising opportunities, led by automation of repetitive work, reducing administrative burden, and analyzing large datasets. (insidehighered.com)

• ROI measurement lags behind adoption

  • Only 13% say their institution measures ROI for work-related AI tools.

  • The article frames ROI as difficult but increasingly important for longer-term investment decisions. (insidehighered.com)

Overlaps between the EDUCAUSE report and the Inside Higher Ed article

• Both highlight a gap between AI use and institutional direction

  • 94% of employees use AI for work, while only 54% know policies or guidelines. (EDUCAUSE Library)

  • Both frame this gap as a governance and risk issue, especially for privacy, security, and data governance. (EDUCAUSE Library)

• Both show high adoption with limited mandate

  • Most people are not required to use AI tools for work, yet most want to keep using them. (EDUCAUSE Library)

• Both present “risk plus opportunity” as the dominant pattern

  • Many respondents select multiple urgent risks and multiple promising opportunities, which supports the “overload” feeling described in the article. (EDUCAUSE Library)

• Both point to shadow AI as a practical problem

  • 56% report using tools not provided by their institution, which increases exposure for privacy, security, and compliance. (EDUCAUSE Library)

• Both flag weak ROI practice

  • Only 13% say their institution measures ROI for work-related AI tools. (EDUCAUSE Library)

Differences in focus and value

• EDUCAUSE provides the full evidence base and structure

  • Survey timing and scale, segmentation, and detailed sections on strategies, policies, risks, opportunities, challenges, use cases, and demographics. (EDUCAUSE Library)

  • Granular task-level use cases and frequency patterns, plus details on what strategies often include (pilots, risk evaluation, encouragement, policy creation). (EDUCAUSE Library)

• Inside Higher Ed adds journalistic interpretation and critique

  • Emphasizes the “disconnect” narrative and quotes the report author on implications. (insidehighered.com)

  • Introduces an external critique about representation and framing, noting the respondent mix includes only 12% faculty and suggesting the enthusiasm signal may reflect staff-heavy use cases and answer-choice effects. (insidehighered.com)

• EDUCAUSE is descriptive and operational, Inside Higher Ed is explanatory and contextual

  • EDUCAUSE supports decision-making with breakdowns you can translate into governance, training, and tool provisioning plans. (EDUCAUSE Library)

  • Inside Higher Ed highlights what the findings mean for campus politics, perception, and how different groups may read the results. (insidehighered.com)

Bottom line you can use

• Use EDUCAUSE for planning and implementation because it gives the task lists, adoption patterns, and strategy components. (EDUCAUSE Library)

• Use Inside Higher Ed to shape communication because it surfaces the core tension, low confidence, and the stakeholder critique you will hear on campus. (insidehighered.com)

Follow us: https://www.linkedin.com/feed/update/urn:li:activity:7418707218562396160

18 яну 2026