AI evidence extraction

The Double-Edged Tool: Student Perspectives on the Ethical Use, Skill Impact, and Pedagogical Adaptation of Generative AI in Higher Education

Mark Kevin Astrero · Research and Advances in Education · 2026

AI-generated evidence extraction, verified across multiple analytical personas. Not a substitute for the peer-reviewed original.

7/10

Relevance

0/4

Quality (LMQS)

Evidence

Citations

0.00

FWCI

This is an AI analysis. Read the peer-reviewed original at the publisher: https://doi.org/10.63593/rae.2788-7057.2025.12.001

Methodology & findings

Study design

Qualitative-phenomenological inquiry using a survey/interview protocol administered via Google Forms to collect self-reported information from college students, with systematic thematic analysis employed to identify patterns, themes, and categories in student responses..

Primary method

Systematic thematic analysis was used to identify patterns, themes, and categories in students' responses. No quantitative statistical methods are mentioned in the abstract.

Main result

Students see GenAI as a double-edged tool that "helps with ideation, summarizing complex topics, and becoming more efficient when writing." However, "it posed a significant problem regarding academic dishonesty and the resultant loss of specific basic competencies, such as analysis, logical reasoning, and independent effort." The study found that students reported "taking measures to mitigate risk, such as conducting extensive fact-checking and regulating their own activities," while many indicated that "clearer standards for disclosing sources were needed."

Reports effect sizes.

Research paradigm

Qualitative-phenomenological

Author conclusions

The authors conclude that "such duality of GenAI requires an institutional response in kind" and specifically suggest that "Higher Education Institutions (HEIs) and faculty move 'beyond a blanket ban approach, toward an integration of use' strategy." They recommend "co-creating clear ethical guidelines; rethinking tasks to become 'AI-resilient' by focusing more on process-based assessment, such as reflection, collaboration, and application in the real world; and shifting instructors' roles to facilitators of deep, process-focused learning."

Risk of bias

Selection bias: Participants self-selected into the study via survey/interview protocol; Self-report bias: Data collected through self-reported information in Google Forms; Geographic limitation: Sample restricted to Filipino higher education institutions; Potential social desirability bias: Students' self-reported measures to mitigate AI use risks; Selection bias: Self-reported data from voluntary survey/interview participants; Potential response bias: Self-selection of students willing to participate; Geographic limitation: Only Filipino college students from Philippine institutions represented; Potential social desirability bias in self-reported ethical behaviors; Self-report bias: Study relied on "self-reported information" which may be subject to social desirability bias; Selection bias: Participants were recruited from "multiple higher education institutions across the Philippines" but sampling method and recruitment criteria not specified; Geographic limitation: Sample restricted to Filipino college students, limiting generalizability; Attrition: No information provided about response rates or completion rates

Open questions raised

The abstract does not explicitly identify specific research gaps or future research directions, though the recommendations for institutional adaptation and pedagogical change suggest areas for future implementation and evaluation research.
The study identifies the need for institutional responses to GenAI integration, including the development of clearer standards for source disclosure, AI-resilient pedagogical approaches, and redefined instructor roles focused on facilitating deep learning rather than content delivery.
The paper identifies the need for institutional strategies beyond blanket bans, clearer standards for source disclosure, and faculty development in facilitating process-focused learning in AI-integrated environments. Future research directions include examining implementation of AI-resilient pedagogical approaches and longitudinal tracking of competency development.

Data: not_statedCode: not_statedExtracted from: pdfAgreement 64%

Explore related topics

Responsible AI Use AI Governance Qualitative