Agentic education · Learning science

Actionable companion · Learning design & outcomes

From AI evidence to responsible practice: Student perceptions of artificial intelligence in higher education: a structural analysis at an Italian university

AI-assisted editorial implementation guide, separate from the original research. Proposed actions are not findings or recommendations attributed to the source authors. No independent peer review of this guide is asserted.

Read original paperSearch paper agentCC BY 4.0

Evidence and context

Title and abstract screening; stage labels describe reported settings, not verified numerical ages. Design labels require full-text appraisal. Reported stage(s): Higher education. Verify the participants, intervention and comparator before applying this study in a new setting.

Original authors’ abstract, reproduced under the linked source license:

Background and aims The rapid advancement of artificial intelligence (AI) is transforming higher education, yet understanding of student perceptions remains limited. This study investigates the structure of student attitudes toward AI and identifies key predictors among Italian university students. Methods A cross-sectional survey was administered to 864 students at the University of Palermo (May–June 2025). The questionnaire examined demographics, AI experience, and attitudes using 10 Likert-type items. Data were analyzed using exploratory factor analysis and confirmatory factor analysis. Results Most students (65.8%) reported superficial AI knowledge, yet 93.6% had used AI applications, predominantly ChatGPT (59.8%). Factor analyses identified two distinct latent constructs: perceived Impact of AI and AI-related Concerns, negatively correlated (r = −0.34, p < 0.001). Perceived Impact was positively predicted by age (β = 0.17), STEM (β = 0.19), Health/Agricultural/Veterinary sciences (β = 0.23), Economics/Law/Social Sciences (β = 0.16), and regular AI use, while negatively predicted by female gender (β = −0.09) and non-use (β = −0.35). AI-related Concerns were positively predicted by female gender (β = 0.21) and non-regular use (never: β = 0.20; occasionally: β = 0.29), and negatively by STEM (β = −0.11) and Health sciences (β = −0.15). Conclusions Student attitudes toward AI reflect two distinct dimensions: opportunity recognition and risk awareness, systematically influenced by gender, discipline, and AI experience. Successful implementation requires tailored approaches addressing gender-specific concerns, discipline-specific needs, and promoting direct AI experience.

What kind of evidence is this?

Supporting AI learning science · Survey / observational study

Stage: Higher education. Screening labels describe reported context and study-design language. They are not a risk-of-bias rating or proof of effectiveness for an age band.

Study methods · source passage

This cross-sectional survey study was conducted at the University of Palermo (UNIPA), Italy, between May and June 2025. The University of Palermo is a comprehensive public research university with approximately 46,000 students across more than 130 degree programs, spanning 16 different academic departments and disciplines. The response rate was 1.88% (864 completed surveys out of approximately 46,000 eligible students).

Limitations and uncertainty · source passage

The findings from this study have several important implications for universities seeking to integrate AI into their curricula. Given the powerful effect of direct AI experience on attitudes, universities should prioritize providing students with supervised opportunities to engage with AI tools in educational contexts. These programs should move beyond superficial tool usage to develop deeper understanding of AI capabilities, limitations, and appropriate applications (Long and Magerko, 2020). The disciplinary divide in perceived impact suggests that generic AI training may not resonate equally across fields. Humanities and social science students need to see AI's relevance to their disciplines through concrete, field-specific applications that align with their professional goals and values. The bidimensional structure of attitudes indicates that addressing concerns does not automatically increase perceived benefits. Successful implementation requires parallel strategies: highlighting AI's educational potential while simultaneously providing concrete solutions to legitimate concerns about privacy, employment, and human interaction. Finally, given the strong effect of AI usage on attitudes, creating communities where students can share experiences, troubleshoot problems, and learn from peers who have successfully integrated AI into their academic work may be more effective than top-down training programs.

Design the system before the pilot

Use this study to inform a learning mechanism or implementation question. It is not direct evidence that autonomous agents improve learning. Keep the proposed agent behavior separate from what the source actually tested.

  • Specify a measurable learning objective, a comparator and a task without tool assistance before adopting a system.

A 90-day district pilot

  1. Days 1–15 · Appraise. Have the curriculum lead and a teacher read the methods and limitations. Record the study design, sample, comparator, outcomes and evidence gaps. Define a local question that the study can help inform.
  2. Days 16–30 · Design. Choose one disciplinary task. Require a first attempt, a documented AI consultation, a critique of the output and an independent explanation. Align the comparison group and scoring rubric with the same objective. Name an accountable owner and set the baseline, success threshold and budget before implementation.
  3. Days 31–60 · Test. Start with a small voluntary cohort, preserve an accessible non-AI route and require teacher review. Use synthetic or de-identified examples; approve data handling and tool terms through district processes.
  4. Days 61–90 · Decide. Blind-score reasoning and accuracy on a new task without AI. Distinguish model examination scores, student perceptions and actual learner gains; only the last directly tests educational benefit. Compare with baseline, document uncertainty and stop or revise if learning worsens, access is unequal or critical errors persist.

The board decision brief

Report the learning problem, evidence quality, participating population, total cost including review time, measured benefit, errors, access differences and remaining uncertainties. Make a dated recommendation to stop, revise or expand. A positive perception survey alone does not establish learning improvement.

Attribution and reuse

Calogero Cammà; Luisa Amenta; Salvatore Battaglia; Ciro Celsa; Giansalvo Cirrincione; Salvatore Contino; Pier Paolo Corso; Salvatore Di Dio; Gabriele Di Maria; Marco Enea; Adriano Fagiolini; Alessandra Ferraro; Mauro Giuffrè; Giosuè Lo Bosco; Domenica Matranga; Roberto Pirrone; Francesco Maria Raimondi; Ilenia Tinnirello; Riccardo Uccello; Marco Vaccaro; Caterina Ventimiglia; Salvatore Vitabile; Massimo Midiri. Student perceptions of artificial intelligence in higher education: a structural analysis at an Italian university. Frontiers in Psychology (2026-06-29). https://doi.org/10.3389/fpsyg.2026.1861001

Companion prepared 2026-11-01. Original authorship and DOI are preserved. The abstract is an attributed source extract; the implementation plan is new AI-assisted editorial material, offered under the same CC BY 4.0 license. This guide has not undergone independent peer review.