AI breakthrough promises relief from digital fatigue in higher education

The findings position ChatGPT not only as a productivity tool but also as a potential “resource” under Job Demands–Resources theory, which frames how work resources can buffer against strain. By reducing cognitive and digital overload, AI appeared to ease the chronic stress linked to long hours of screen exposure and preparation work in online and blended teaching environments.

A team of Latin American researchers has found that generative AI can significantly reduce digital fatigue and screen time among university instructors, offering a promising tool to improve well-being in higher education. The study is one of the first experimental investigations into how AI affects faculty workload and digital stress.

The paper, Can ChatGPT Ease Digital Fatigue? Short-Cycle Content Curation for University Instructors, was published in Education Sciences. Using a single-case experimental design, the researchers tested whether replacing full readings with AI-generated summaries and discussion prompts could lighten the digital burden faced by instructors. Their results show notable reductions in fatigue, daily screen time, and workload pressures, alongside high acceptance of AI as a teaching support tool.

Can generative AI really cut digital fatigue in higher education?

The study was conducted at a private Peruvian university with eight lecturers from different disciplines. Over eight weeks, participants followed an AB–AB experimental sequence alternating between traditional lesson preparation and AI-supported content curation. In the intervention phases, instructors received daily ChatGPT outputs consisting of a concise 200-word summary of assigned readings and three open-ended discussion questions.

The impact was measurable. Using the validated Digital Fatigue Scale (FDU-24), the researchers found that fatigue scores fell by roughly 22 to 25 percent during AI-supported phases compared to baseline. This improvement was consistent across all lecturers, regardless of subject area.

The findings position ChatGPT not only as a productivity tool but also as a potential “resource” under Job Demands–Resources theory, which frames how work resources can buffer against strain. By reducing cognitive and digital overload, AI appeared to ease the chronic stress linked to long hours of screen exposure and preparation work in online and blended teaching environments.

How much screen time did instructors save with ChatGPT?

Beyond reducing fatigue, the study reported striking gains in time management. Daily logged screen exposure fell by an average of 122 minutes, nearly two hours per day, representing a 29 percent reduction compared to baseline.

This decline was not the result of cutting corners. Instead, ChatGPT’s concise summaries and ready-to-use discussion questions allowed instructors to focus on teaching strategy rather than sifting through extensive reading materials. The researchers emphasize that instructors did not bypass content but engaged with it more efficiently, preserving pedagogical depth while cutting unnecessary hours of digital strain.

Treatment fidelity was high, with instructors adhering to the intervention protocols at a rate of 96 percent. No adverse effects were recorded, suggesting that AI support was both practical and safe to integrate into daily teaching routines.

Effect sizes were also significant. Using metrics such as Tau-U and Cohen’s d, the researchers demonstrated that the intervention produced strong, replicable results across multiple cycles. This strengthens confidence that the observed reductions in fatigue and screen time were not coincidental but causally linked to the ChatGPT intervention.

Are instructors willing to embrace AI in their workflows?

Another key question explored by the study was whether faculty members would accept AI as a long-term teaching aid. At the end of the intervention, all participants completed a technology acceptance survey. Average scores were high: 6.2 out of 7 for perceived usefulness and 6.4 out of 7 for ease of use. Importantly, every lecturer rated both measures at least 5 out of 7, reflecting broad acceptance across the group.

These findings contrast with common concerns about faculty skepticism toward AI. Instead, the study suggests that when AI is framed as a practical, workload-reducing tool rather than a replacement for academic expertise, it is readily embraced. The results also highlight that generative AI does not need to replace traditional methods but can complement them through targeted, short-cycle interventions.

Still, the authors caution that their study was small-scale and limited to a single institution. While the results are promising, larger multi-site randomized trials are needed to test generalizability, long-term effects, and cost-effectiveness. The researchers also stress the importance of maintaining academic integrity, ensuring that AI tools remain transparent and do not erode critical engagement with scholarly content.