Scholars have increasingly focused on the use of GenAI in higher education since its inception, little is known

The gap

Evidence profile

Sourced from the abstract and stated research gap and recommendations and inline gaps of the source papers, classified as general, drawn from work published between 2024 and 2026, spanning 6 journals. Those papers have been cited 86 times in total.

Supporting evidence — 6 representative gaps

• Using generative artificial intelligence/ChatGPT for academic communication: Students' perspectives (2024) · International Journal of Applied Linguistics · cited 86× · doi

  While scholars have increasingly focused on the use of GenAI in higher education since its inception, little is known about how key higher education stakeholders, particularly students, perceive its impact on teaching and learning within the context of academic communication, an area central to students' development of transferable skills and literacy competencies yet heavily influenced by the technology.

  generalabstractevidence 5/5
  Keywords: higher education students scholars increasingly focused genai inception little known stakeholders particularly perceive impact teaching
• Multinational machine learning investigation of whether generative AI interactions augment or substitute cognitive skills and study efficiency in global higher education (2026) · Discover Artificial Intelligence · doi

  The paper identifies a gap in the empirical evidence base guiding institutional policy, pedagogical design, and student practice in the context of GenAI adoption. The study highlights the need for research on the impact of GenAI on cognitive skills and study efficiency in higher education. The paper notes that recent studies have focused on the benefits of GenAI, but have neglected the potential costs, such as cognitive offloading.

  generalstated research gapevidence 5/5
  Keywords: paper identifies gap empirical evidence base guiding institutional
• “AI Should Help them Learn, Not Learn for Them”: University Staff Perspectives on the Role of Generative AI in Education (2026) · Journal of University Teaching and Learning Practice · doi

  To effectively navigate the integration of GenAI into higher education, a comprehensive institutional strategy is essential. This strategy should include the development of clear guidelines, targeted training, and structured support for both staff and students (Fawns, 2022; Kutty et al., 2024; Sharples, 2023; Siemens, 2005). Within this context, a broader pedagogical discussion around GenAI use in learning and teaching is necessary, extending beyond resource provision and assessment design. A rethink of pedagogy that addresses the dynamic changes in educational practice is required. Importantly, these theoretical perspectives help illuminate the staff concerns identified in the findings, particularly those relating to integrity, workload pressures, uncertainty about appropriate use, and shifting expectations of academic labour. When viewed through a broader post digital and sociotechnical lens, these concerns become more intelligible, especially the tensions evident in staff reflections about assessment stability, confidence in working with GenAI, and the need for clearer pedagogical direction. This framing directly informs the strategies proposed below. For example, drawing on entangled pedagogy (Fawns, 2022; Siemens, 2005) we might consider how a new cooperative social digital learning model (Sharples, 2023) could shape our practice. This model could incorporate elements of self-regulated digital learning (Jin et al., 2023), co- regulated learning (Lodge et al., 2023) and explore the broader conversations of educational pedagogy and digital collaboration (Sharples, 2023). We should recognise AI in assessment as an ever-developing condition to be navigated (Corbin et al, 2025) whilst firmly prioritising accessibility and inclusion (Kelly et al., 2023). Building on this educational approach, an institutional GenAI educational framework (Hillier, 2023) would be beneficial including a whole institution program level review of our assessment practices (TEQSA, 2024). This would include a rethink of our curriculum approach to incorporate programmatic approaches (Baartman & Quinlan, 2024), evaluative assessment design (Bearman et al., 2024) and visible learning integration (Bearman et al., 2024) ensuring our learning and teaching strategies incorporate GenAI literacy and student social collaboration at its core. Practical strategies for staff in mitigating AI risk through visible assessment design and an AI integrated approach to assessment regimes would support academics currently navigating this challenging landscape. We need to be mindful of the complex interplay between efficiency and quality, enthusiasm and risk. Whilst universities face huge tensions in transforming educational practice given this complexity, we need to acknowledge the ‘wicked problem’ of AI which we must all navigate rather than solve (Corbin et al., 2025).

  generalrecommendationsevidence 5/5
  Keywords: assessment learning genai educational staff digital sharples broader design pedagogy practice need strategies incorporate approach
• Generative Artificial Intelligence in Tertiary Level Education in Bangladesh: Practices, Benefits, Challenges, and Prospects (2026) · Journal of Information Technology Education Research · doi

  for Researchers This study contributes empirical evidence on the adoption of GenAI in a South Asian tertiary education context, enriching the body of knowledge on technology acceptance, digital pedagogy, and GenAI in education policy. By revealing the pictures of relevant variables of Generative Artificial Intelli- gence in Education (GenAIEd) in a unique context, such as Bangladesh, the findings have implications for similar situations. They can inform others about possible challenges and the usefulness of GenAIEd. Teachers and students are both moderately familiar with GenAI. The teach- ers primarily use it to prepare courses and materials, while students sporadi- cally engage with GenAI, mainly for academic problem-solving, and they em- phasize its role in personalized, learner-centered learning. GenAI familiarity is found to be a strong predictor of usage frequency. However, teachers ex- press concerns about the reliability of GenAI, ethical implications, and the potential for deskilling. While the benefits and usefulness dominate, possible challenges and threats are marginally associated with the future adoption and use of GenAI. This finding is unique because, despite the overpowering ‘ease of use’ of the TAM model, ‘benefits or usefulness’ of the TTF model, chal- lenges, and threats have been found as catalysts for GenAI adoption. Practitioners are to utilize GenAI to support, rather than replace, their teach- ing expertise. They should also encourage students to strike a balance be- tween GenAI-assisted learning, critical thinking, and independent work. Fur- thermore, the institutions should introduce guidelines to ensure the ethical use of GenAI and academic integrity. Researchers should explore the longitudinal effects of GenAI adoption on learning outcomes and skill development. They can also conduct compara- tive studies across different universities and disciplines. Investigating the role of GenAI in inclusive education and support for learners from disadvantaged backgrounds also demands research focus. Impact on Society The findings highlight how GenAI can transform higher education in Bang- ladesh and similar contexts. It shows the importance of addressing the risks of overreliance and the unethical use of GenAI for effective learning. A bal- anced adoption could strengthen human–technology collaboration in educa- tion. On the other hand, it has revealed the aspects of GenAI, preferred by educators, that AI developers should consider.

  generalrecommendationsevidence 5/5
  Keywords: genai adoption education learning usefulness students researchers context technology genaied unique implications similar possible challenges
• The Affective Impact of Generative Artificial Intelligence on Instructors of First-Year Writing Courses (2026) · Teaching & Learning Inquiry The ISSOTL Journal · doi

  In response to our findings, we offer three preliminary recommendations for institutions and departments to address the affective impacts of GenAI on FYW instructors. The first involves a call to action to 1) provide clearer guidelines on addressing potential cases of unauthorized use of GenAI in written submissions, thereby reducing added labour and allocating some protections to vulnerable faculty, and 2) deliver guidance on conducting exploratory meetings with students in suspected cases of unauthorized GenAI use. For example, workshops on standards of evidence in academic codes of conduct and how they relate to suspected unauthorized GenAI use would alleviate the burden that Caroline identified in her attempts to “wade through” the “big, long, legal material” of institutional academic integrity policies in order to determine what approach to take in cases of suspected GenAI infractions. She proposed that a “forum like a lecturer question and answer” may better suit the needs of faculty. A second recommendation is that institutions expand the range of supports offered to faculty in the wake of GenAI, specifically addressing the emotional and interpersonal pedagogical effects of the technology. Whereas workshop sessions on integrating GenAI tools into teaching and assignment design proliferate, guidance on addressing issues of trust-building in the TSR would constitute valuable additions. A final recommendation comes from an unexpected finding: Five out of eight participants expressed gratitude for the chance to discuss the emotional consequences of the introduction of Paxton, Amanda, Phoebe Kang, and Ashley Yim. 2026. “The Affective Impact of Generative Artificial Intelligence on Instructors of First-Year Writing Courses.” Teaching & Learning Inquiry 14: 1–17. https://doi.org/10.20343/teachlearninqu.14.19 12 THE AFFECTIVE IMPACT OF GENERATIVE ARTIFICIAL INTELLIGENCE ON INSTRUCTORS OF FIRST-YEAR WRITING COURSES GenAI, sometimes voicing surprise at the “cathartic” (Gareth) effect of the interview. Filomena joked, “I feel like I should pay you, because you’re listening to me vent.” The opportunity to speak with colleagues about lived experience in this turbulent moment could be made available at minimal cost to the institution and would serve the collective good. We therefore recommend formal support for communities of practice in which instructors can share their experiences with each other for mutual support and acquire guidance from experts. This suggestion aligns with the findings of Kohnke, Di Zou, and Moorhouse (2024), who identified the value of communities of practice among English language instructors as they navigate the emergence of GenAI. FUTURE DIRECTIONS FOR RESEARCH In future research, we aim to expand our dataset by including a larger and more diverse group of writing instructors from various post-secondary institutions. This would allow for a deeper exploration of how di

  generalrecommendationsevidence 5/5
  Keywords: genai instructors institutions affective first addressing cases unauthorized faculty guidance suspected writing academic identified like
• The illusion of cognitive independence in the age of generative AI: unpacking GenAI literacy and learner profiles (2026) · Frontiers in Education · doi

  Additional work is needed to explore how these dynamics vary across disciplines, educational contexts, and levels of GenAI integration. First, they suggest that focusing on the extent of GenAI use is insufficient; greater attention should be given to the quality of use and how learners combine monitoring with reliance. an understanding of and Second, the findings indicate that critical-ethical awareness alone is insufficient to ensure responsible use.

  generalinline gapsevidence 5/5
  Keywords: genai insufficient additional needed explore dynamics vary across disciplines educational contexts levels integration first suggest

Questions about this gap