Author: NCFDD
Artificial intelligence has become a regular part of conversations about productivity in higher education. From drafting emails to organizing research materials, AI tools offer opportunities to complete certain tasks more efficiently. For faculty balancing teaching, research, service, and administrative responsibilities, the possibility of getting some time back is appealing.
Yet the relationship between AI and faculty workload is proving more complicated. While some tasks are becoming easier, faculty are also taking on new responsibilities as they adapt to AI’s growing presence in higher education. Redesigning assignments, verifying student work, addressing academic integrity concerns, and establishing expectations for AI use all require time.
NCFDD’s Faculty Perspectives on AI survey of 1,659 faculty members and academic administrators offers a closer look at how those demands are affecting academic work. Among respondents, 35% say AI has increased their workload, compared with 21.7% who report a reduction.
Those findings raise an important question for faculty and institutional leaders: How much time is AI actually saving when the additional work it creates is taken into account?
Where AI can save faculty time
There are meaningful opportunities for AI to make parts of academic work more efficient. Faculty can use these tools to organize information, review lengthy documents, prepare routine communications, and develop starting points for certain administrative or scholarly tasks.
For someone managing a demanding workload, even relatively small improvements can be valuable. Streamlining a recurring task may create more room for teaching preparation, research, student engagement, or other priorities.
Faculty who use AI regularly are also more likely to report these benefits. Among the most confident AI users in NCFDD’s survey, 54% say AI has reduced their workload, while 11% report an increase.
However, the time required to review and refine AI-generated work can affect those savings. Outputs may require fact-checking, substantial editing, or additional research before they are useful. Faculty working with sensitive information or specialized disciplinary content also have to consider accuracy, privacy, and the appropriate role of AI in their work.
The time saved on a particular task is therefore only part of the calculation. Faculty still have to manage the broader responsibilities of teaching and scholarship, including those that are changing because students and institutions are using AI.
How AI is adding to faculty teaching workloads
Some of the most substantial changes are happening in course design and assessment.
Faculty have long invested time in developing assignments that help students practice new skills, apply knowledge, and demonstrate understanding. Generative AI has complicated that process by making it possible to produce polished written work, functioning code, and other completed assignments with relatively little evidence of how much the student understands.
As a result, many instructors are reconsidering established assessment practices.
NCFDD found that 86.9% of survey respondents have changed at least one assessment practice in response to AI. More than half are redesigning assignments to require critical thinking, while 55.7% have introduced more in-class or oral assessment.
These changes can be substantial. An instructor who previously relied on a written assignment may now incorporate draft submissions, individual reflections, classroom activities, or oral presentations. Each change can require new instructions, grading criteria, scheduling considerations, and additional conversations with students.
That work can extend across an entire course. As one humanities faculty member put it:
“I’ve had to redesign almost every assignment and assessment so it can be completed and evaluated in the context of an AI environment. So. Time. Consuming.”
Redesigning a course can be worthwhile, particularly when the changes help faculty better understand what students are learning. But that investment of time has to fit somewhere within an existing workload. And as AI capabilities continue to evolve, faculty may need to revisit approaches they recently developed.
The additional work of verification and academic integrity
Assessment redesign is only one source of added teaching responsibilities.
Faculty are also navigating uncertainty about how students produce their work and what that work demonstrates. In NCFDD’s survey, 77.6% of respondents say AI makes it harder to assess what students actually know. Meanwhile, 80% lack confidence in AI or plagiarism detection tools.
These concerns can lead faculty to spend additional time reviewing student submissions, examining sources, asking follow-up questions, and creating opportunities for students to demonstrate their understanding.
Academic integrity investigations can be particularly demanding. Suspected misconduct may require gathering documentation, meeting with students, consulting institutional policies, and participating in formal review processes.
For some faculty, that work is taking time away from other teaching responsibilities.
“My designated teaching time is no longer spent providing meaningful feedback to students or developing exciting new activities,” explained one humanities faculty member. “It is now spent playing gotcha and putting together misconduct cases. It takes on average 8 hours for one misconduct case.”
Those hours come out of an already full workload, leaving less time for the feedback, course development, and student engagement that faculty want to prioritize.
The implications also extend to faculty who do not personally use AI. They may still need to adjust assignments, address student questions, and respond to its presence in their classrooms. Among survey respondents who intentionally do not use AI, 62.5% report that their workload has increased.
AI policies are creating new decisions for faculty
Developing clear expectations around AI use is another responsibility that has become part of course preparation.
Faculty must consider when AI assistance is appropriate, what students should disclose, and which skills students need to demonstrate independently. Those expectations can vary considerably depending on the discipline, course level, and purpose of an assignment.
The survey found that 71.2% of respondents set their own AI policies for students, while only 8.5% follow a university-wide policy.
Faculty need room to make decisions that reflect their disciplines and learning objectives. At the same time, developing policies independently can require considerable effort, particularly when institutional guidance is limited.
An instructor may need to explore how AI tools perform on common course assignments, consider the implications for learning, revise syllabus language, and explain appropriate use to students. Those decisions may also need to be revisited as technology changes.
Across an institution, similar questions may be considered separately by dozens or hundreds of instructors. Providing shared resources and opportunities for faculty collaboration could reduce duplicated effort while allowing appropriate differences among courses.
Understanding AI and faculty workload beyond time savings
The workload conversation becomes especially important when considering how these changes affect faculty well-being.
NCFDD found that 64.9% of respondents describe AI as another burden on an already full plate, while 60.2% say it is increasing their stress.
The relationship between workload and stress is particularly striking. Among respondents whose workload increased because of AI, 89.4% also reported increased stress. Among those whose workload decreased, that figure was 19.7%.
These findings show a strong association between reported workload changes and faculty experiences, although they do not establish a causal relationship.
They also highlight why productivity discussions need to consider more than the efficiency of individual tasks.
One social sciences faculty member described an experience that captures both sides of the workload question:
“I use AI as a glorified Google. You could say it has saved me time in that. But it has also increased the time I spend on teaching because I had to redesign my courses and move to in-class hand-written exams which are harder to grade.”
Even when AI helps streamline certain responsibilities, changes elsewhere in teaching can absorb those gains. The overall effect depends on how much additional work faculty take on and whether they have the time and resources to manage it.
For institutions hoping to support more sustainable academic careers, that broader view of faculty workload matters.
What institutions can do to support faculty time
As AI continues to reshape teaching and scholarly work, institutions have an opportunity to consider how their expectations affect faculty workload.
Professional development is part of that effort, but faculty also need time and resources to apply what they learn. A workshop may introduce new tools or assessment strategies, while implementing those ideas across an entire course can require a much larger investment.
Institutions can take several practical steps to help faculty manage these responsibilities:
- Provide dedicated time for course and assessment redesign. Recognize that adapting existing teaching practices requires preparation, experimentation, and revision.
- Develop shared guidance and adaptable resources. Offer clear institutional principles, sample policy language, and assessment resources that faculty can tailor to their courses.
- Create opportunities for faculty collaboration. Departmental conversations and peer learning can help instructors share approaches and avoid repeatedly solving similar problems independently.
- Consider AI-related responsibilities in workload planning. Account for the time spent on verification, student questions, academic integrity processes, and ongoing professional development.
These approaches can help institutions make AI-related changes more manageable while preserving faculty judgment about their teaching and scholarship.
AI may create valuable efficiencies in academic work, and some faculty are already experiencing meaningful time savings. At the same time, the growing demands surrounding its use deserve attention.
As institutions continue exploring AI’s potential, understanding its effect on faculty time will require looking at the full scope of academic work, including the responsibilities that are emerging and changing alongside the technology.
Read NCFDD’s Faculty Perspectives on AI survey report for more findings on how AI is affecting faculty workload, teaching, assessment, and institutional support.
