Science AI Journal
Social Sciences · Research topic

Open research questions in Artificial Intelligence in Education

635 gap statements mined from Artificial Intelligence in Education papers in our 7.9M-paper local library, which holds 2,775 papers on the topic — drawn mostly from each paper's own stated research gap, future-work, challenge and limitation notes, and its abstract. The ones listed below are a selection still marked open; each names the study that raised it, with a DOI link where the paper has one.

Choosing where to publish on Artificial Intelligence in Education? See the ranked Social Science journals.

What the literature leaves open

  • Educators, students, and institutions need to be extremely cautious when using GenAI tools for educational purposes and ensure they understand the fundamentals of GenAI systems, including capabilities, limitations, and privacy. They should seek this information on the tools’ websites and avoid tools that do not provide it.

    Hard to find, harder to understand: examining transparency in educational generative AI · 2026 · DOI

  • Moreover, the study does not include longitudinal follow-up, leaving open questions about how these reflexive practices evolve once students enter professional fields or confront AI in other domains of their academic lives.

    Romanticising the algorithm: power dynamics and ideological implications in human–AI collaboration for higher education · 2026 · DOI

  • To address this challenge, institutions may consider equipping staff with the resources to lead activities that socialise students to critically engage with GenAI outputs (e.g. Gholiagha et al. 2025). Staff-led integrations of GenAI will focus on recognising its potential to enhance learning through greater inclusion and accessibility, while also acknowledging the ethical considerations underpinning its use.

    Pedagogical conundrum, assessment anxiety, and employability dilemma: a qualitative exploration of staff perspectives on the great GenAI quandary in higher education · 2026 · DOI

  • Rather, it is to recognise that learning requires spaces where answers are not immediately available, where questions linger, and where learners AI & SOCIETY are permitted to remain uncertain.

    Reflections on the Always-Answering Classroom: AI, Uncertainty, and Pedagogical Silence · 2026 · DOI

  • 2 suggests that the instrument may have two factors, but it has not been confirmed by confirmatory factor analysis (CFA) with an independent sample.

    Teachers’ perceptions of artificial intelligence integration in education: a multi-campus cross-sectional study from Riphah International University, Pakistan · 2026 · DOI

  • Future studies should examine the approach with larger cohorts and across different physics topics and educational contexts.

    Do-It-with-AI Framework for Invention: A Kundt’s Tube Case Study · 2026 · DOI

  • Future research should examine similar dimensions in less structured tasks to determine the extent to which these patterns persist when fewer pedagogical criteria are made salient in advance. Future research should examine the relationship between pedagogical intentions, prompt characteristics, AI-generated outputs, subsequent human revisions, and the quality of the final educational resource.

    Prompt Design as AI-Mediated Pedagogical Performance in Initial Teacher Education: A Case Study in Foreign Language Teaching Methodology · 2026 · DOI

  • Together, these complementary directions would address key limitations of the present study and provide a more comprehensive basis for examining TouriBot across methodological, technical and educational dimensions. The findings are further limited by the use of self-reported data, which may be vulnerable to response biases such as social desirability and acquiescence.

    Design and Multi-Stakeholder Evaluation of the Perceived Pedagogical Value and Usability of an AI-Supported Educational Chatbot for Primary Education · 2026 · DOI

  • Where possible, institutions should also develop device-loan or device-support programs for students who lack suitable personal devices. Reduce the affordability barrier: Institutional digital access strategies should recognise that students who rely on mobile data may be unable to use GenAI consistently. Expanded campus connectivity and other forms of supported digital access could reduce this barrier.

    Generative AI and Digital Inequality in South African Higher Education: Students’ Experiences of Access and Meaningful Use at a Rural University · 2026 · DOI

  • A smartphone can provide access to an AI platform, but a weak network, insufficient data, or reliance on intermittent campus Wi-Fi can make that access fragile.

    Generative AI and Digital Inequality in South African Higher Education: Students’ Experiences of Access and Meaningful Use at a Rural University · 2026 · DOI

  • Rather than relying on a single theoretical lens, the framework integrates disability studies, learning sciences, motivational psychology, learning analytics, and AI governance to address the complex challenges of supporting students with learning difficulties in higher education.

    Toward an AI- and learning-analytics-enabled inclusive support framework for students with learning difficulties in Kuwait's higher education institutions · 2026 · DOI

  • For these teachers, AI was not characterized as a replacement for teachers or human relationships, but rather as one resource within a broader pedagogical and relational framework.

    AI-supported activities for social inclusion: pre-service teachers’ perceptions of opportunities and challenges in special education · 2026 · DOI

  • Building on the gaps identified in this review, future research should prioritize expanding the geographic and educational-level scope of inquiry beyond Java-based higher education institutions.

    AI Integration in Indonesian Education: Trends, Outcomes, and Challenges - A Comprehensive Review · 2026 · DOI

  • At the same time, because formal HTMT or latent-factor correlation statistics were not included in the original analysis, strict statistical discriminant validity among the three highly related dimensions should be examined further in future validation studies.

    Development and validation of the AI cognitive offloading scale for Chinese college students · 2026 · DOI

  • This deployment was a pilot: one course, one teacher, one institution, and a single cohort of 33 students; no data were collected or analyzed, no reflective output examined, and there was no comparison group.

    What a teacher cannot delegate: a pedagogy-first framework for integrating conversational AI agents across the simulation cycle · 2026 · DOI

  • Limited availability of physical evidence and access to morgue facilities. Logistical costs and need for standardized assessment tools. Heterogeneity of the evidence and methodological limitations. Need for adaptation of the educational infrastructure to the evolving field of forensic science.

    Artificial intelligence applications in forensic science and their implications for education and training · 2026 · DOI

  • Implementation guide/measurable indicator 12 Adopt a hybrid human-in-the-loop moderation model with a clear Where AI is deployed to support evaluation, feedback, or grading, AI-generated student contestation pathway when AI-supported evaluation is used.

    ChatGPT as an AI-based assessment method in library information science education: investigative study in the information studies department · 2026 · DOI

  • Implementation guide/measurable indicator 1 2 3 Develop a departmental AI assessment policy for Information Studies The department should prepare one shared policy template that defines permitted, courses.

    ChatGPT as an AI-based assessment method in library information science education: investigative study in the information studies department · 2026 · DOI

  • Based on these findings, the following recommendations have been developed.

    Ecological Invisibility in Classroom AI Use: Rethinking Pedagogical Efficiency, Sustainability and Teacher Agency · 2026 · DOI

  • Station Rotation Items 1–3 46.2/53.8/ 13 (A, 20.11.25) 76.9% Fishbowl (C, 08.04.26) 5 items 75.0/66.7/66.7/ 12 100.0/75.0% aTwo additional single-respondent Gruppenpuzzle records (10.10.25 and 11.10.25, N = 1 each, evidently make-up submissions from the same Group A cohort) are excluded from this table’s percentages but included in Table 1’s total N = 28.

    Redesigning collaborative teaching for human–AI partnership: a phase-based AI-constraint design principle in higher education · 2026 · DOI

  • AI self-efficacy showed a positive serial indirect association through information utilization and perceived efficiency, while its adjusted direct associations and shorter PWE-only indirect route were inconclusive.

    Responsible GenAI Augmentation and Cognitive Agency in Higher Education · 2026 · DOI

  • Although scheduled classroom observations and teacher discussions were used to monitor adherence to the instructional protocol, no standardised fidelity checklist or rating scale was administered, no quantitative fidelity scores were generated, and no independent second rater was used to establish inter-rater reliability.

    Teacher-Mediated, Not Learner-Facing: A Bayesian Quasi-Experimental Study of Agentic AI Support in Grade 10 Functions and Graphs · 2026 · DOI

  • We examined intrinsic motivation as the sole moderator and only tested its moderating effect on the pathway from AI trust to career confidence, so the boundary conditions operating at other stages of the serial mediation model remain unclear; future research could investigate additional moderators…

    The association between generative AI acceptance and college students’ career preparation: the serial mediating role of AI trust and career confidence and the moderating role of intrinsic motivation · 2026 · DOI

  • Expected outcome: Academics who confidently use AI to enhance inclusion while preserving educational quality and academic integrity. 8.3.4 Layer 4.

    Toward an AI- and learning-analytics-enabled inclusive support framework for students with learning difficulties in Kuwait's higher education institutions · 2026 · DOI

  • Studies that compared GenAI with instruction without AI almost always reported positive outcomes, while the few comparisons between GenAI designs gave mixed results, with some support for withholding direct answers and adding metacognitive scaffolds.

    How Does Generative AI Support Inquiry-Based Learning? A Systematic Review · 2026 · DOI

Most-cited papers in Artificial Intelligence in Education

Most recent work

Find a gap in your own Artificial Intelligence in Education sub-topic

This page shows what the Artificial Intelligence in Education literature already flags as unresolved. To narrow it to your specific question, search the Research Gap Finder: the search is free with a free account and lists the papers closest to your topic first. Unlocking that topic (50 credits, charged once) fills the comparison table from our 7.9M-paper local library and writes the gaps from its rows.

Open the Research Gap Finder →

Related topics in Social Sciences

Legal Issues in Turkey
2,143 open questions
Criminal Justice and Corrections Analysis
2,104 open questions
Teacher Education and Leadership Studies
2,009 open questions
Impact of Technology on Adolescents
1,855 open questions
Higher Education Research Studies
1,841 open questions
Ethics and Social Impacts of AI
1,768 open questions
Global Educational Policies and Reforms
1,735 open questions
Urban Transport and Accessibility
1,724 open questions
Open questions raised by papers listed on this page include: (1) The study does not include longitudinal follow-up, leaving open questions about how these reflexive practices evolve once students enter professional fields or confront AI in other domains of their academic lives (“Romanticising the algorithm: power dynamics and ideological implications in…”, 2026); (2) Future studies should examine the approach with larger cohorts and across different physics topics and educational contexts (“Do-It-with-AI Framework for Invention: A Kundt’s Tube Case Study”, 2026); (3) Future research should examine similar dimensions in less structured tasks to determine the extent to which these patterns persist when fewer pedagogical criteria are made salient in advance (“Prompt Design as AI-Mediated Pedagogical Performance in Initial Teacher…”, 2026). 635 gap statements have been mined from Artificial Intelligence in Education papers in our 7.9M-paper local library, which holds 2,775 papers on the topic; the gaps come from whichever of those papers state one. They are mostly the research gaps the authors state and the papers' abstracts, plus future-work, limitations and challenges passages. The ones listed below are a selection still marked open; each names the study that raised it, with a DOI link where the paper has one, so you can read the original claim in context.

Tools for your next paper

Gaps & Literature SearchFind open questions in any field — 1,100,000+ open gaps mined from our own library.Pre-CheckIs your paper ready? Acceptance odds for top, mid and open/emerging journals, in seconds.Journal FinderA ranked shortlist of target journals from a 17,500-venue index.Duplicate Publication CheckerCheck for prior publication across 7 sources and our local library.Citation GeneratorFreeA DOI or title → APA, MLA, Vancouver, Chicago, BibTeX, RIS. No signup.Graphical Abstract MakerFreeTurn your findings into a colorblind-safe graphical abstract.

How Pre-Check worksHow AI Review works

Compare the category — Honest roundups of the AI research tools, ours listed alongside the alternatives.

Command palette

Jump anywhere, run any action.