AI evidence extraction

GENERATIVE ARTIFICIAL INTELLIGENCE FOR MICROLEARNING CONTENT DEVELOPMENT: OPPORTUNITIES, CHALLENGES, AND PEDAGOGICAL FRAMEWORKS

Gregorius Punto Aji, Benedecta Indah Nugraheni · International Journal of Education and Social Science Research · 2026

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

6/10

Relevance

1/4

Quality (LMQS)

Evidence

Citations

0.00

FWCI

This is an AI analysis. Read the peer-reviewed original at the publisher: https://doi.org/10.37500/ijessr.2026.9101

Methodology & findings

Study design

Narrative literature review with structured and iterative searches.

Sample

N = 20, 1 group

Primary method

Thematic analysis with deductive (opportunities, challenges, frameworks) and inductive coding. No statistical tests or quantitative synthesis methods employed.

Main result

The study found that "Gen AI supports key development tasks, including learning objective formulation, outlining and storyboarding, multimodal content production, and micro-assessment generation." Additionally, "Opportunities include faster and more scalable production, expanded multimodal formats, and stronger support for personalization and accessibility."

Reports effect sizes.

Research paradigm

Interpretivist/qualitative synthesis

Author conclusions

The authors conclude that "The reviewed literature points to an eclectic, theory-informed approach (grounded in cognitive and multimedia learning principles, instructional design models, microlearning-specific frameworks, and inclusive design) to guide responsible Gen AI use."

Risk of bias

Single database search (Google Scholar only) - potential publication bias and missing gray literature; Recency bias - focus on 2023–2025 studies may overrepresent emerging claims; Selection bias - authors determined inclusion/exclusion criteria without pre-registration; Narrative review design - no systematic quality assessment of included studies; Selection bias: Limited to Google Scholar searches, which may not capture all relevant databases or grey literature; Publication bias: Focus on 2023–2025 core studies may miss relevant earlier work; Reviewer bias: Narrative review methodology is susceptible to subjective interpretation and selection of themes; Language bias: Only studies available through Google Scholar searches may be included, potentially excluding non-English publications

Limitations

The review acknowledges that "Literature was identified through structured and iterative Google Scholar searches (2018–2025)" which relies on a single search platform rather than multiple databases
Additionally, the scope is limited to "20 core studies published in 2023–2025" with earlier sources used only for conceptual background, potentially missing relevant earlier work and limiting the breadth of evidence synthesis.

Open questions raised

The paper identifies needs for: (1) empirical evidence on long-term learning outcomes from Gen AI-developed microlearning content; (2) frameworks addressing ethical concerns including bias, copyright, transparency, and data privacy; (3) guidance on tool/workflow integration; (4) research on balancing fragmentation risks with personalization benefits.
The review identifies gaps related to: (1) operationalizing responsible Gen AI use in microlearning; (2) understanding long-term pedagogical impacts of Gen AI-developed microlearning; (3) addressing ethical concerns including bias, copyright, transparency, and data privacy in Gen AI content generation.
The paper implicitly identifies gaps in understanding how to responsibly integrate Gen AI with microlearning while addressing ethical concerns, bias mitigation, copyright issues, and maintaining learning coherence across fragmented content units.

Data: not_statedCode: not_statedExtracted from: pdfAgreement 59%

Explore related topics

AI Manuscript Writing Doctoral Training Literature Review