AI evidence extraction

Generative artificial intelligence-driven adaptive learning for sustainable, personalized, and resilient education systems

Manjunath Munenakoppa, Nitin Liladhar Rane, Jayesh Rane, Shreeshail Heggond · International Journal of Applied Resilience and Sustainability · 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/10.70593/deepsci.0202011

Methodology & findings

Study design

Systematic literature review using PRISMA methodology examining existing peer-reviewed and academic literature on generative AI in adaptive learning systems across K-12 education, higher education, professional development, and lifelong learning contexts..

Primary method

Systematic review methodology (PRISMA); no quantitative synthesis or statistical pooling is reported. This is a narrative synthesis review.

Main result

The findings indicate that "generative AI technologies, such as large language models, multimodal AI systems, and intelligent tutoring agents, are transforming the mechanisms of delivering education" with demonstrated improvements in "real time content creation and the personalized feedback system, automated evaluation systems and curriculum modification through intelligence." Learning performance is improved through "adjustable difficult levels, generation of contextual content, and customized learning processes, as well as in the context of environmental sustainability through minimized physical resource use."

Reports effect sizes.

Research paradigm

Interpretivist/Critical realist

Author conclusions

The authors conclude that "based on the generative AI-controlled systems, learning performance is improved in the form of adjustable difficult levels, generation of contextual content, and customized learning processes, as well as in the context of environmental sustainability through minimized physical resource use." They further note that "opportunities that are in the offing include hybrid human-AI instruction frameworks, emotion cognizant adaptive frameworks, and blockchain incorporated credentials validation."

Risk of bias

Publication bias (not explicitly assessed); selection bias inherent to literature review methodology; potential researcher bias in study selection and interpretation; Literature review bias (selection of included studies not fully detailed); Potential publication bias (only published studies typically included in systematic reviews); Lack of quality assessment tools specified for included studies; Narrative synthesis may introduce interpretation bias; Publication bias (only published literature reviewed); Selection bias in literature search strategy; Potential bias toward positive AI outcomes in published studies; Geographic and linguistic bias in literature availability

Limitations

The review identifies critical issues including "algorithmic bias, data privacy problems, digital equity differences, and complexities in pedagogical integration" as crucial challenges that need to be addressed in the implementation of generative AI-driven adaptive learning systems.

Open questions raised

The review identifies future research directions including: development of hybrid human-AI instruction frameworks, emotion-cognizant adaptive frameworks, and blockchain-incorporated credentials validation systems. Gaps also include addressing algorithmic bias, data privacy concerns, and digital equity challenges.
The review identifies several future research directions: hybrid human-AI instruction frameworks, emotion-cognizant adaptive frameworks, blockchain-incorporated credentials validation, and addressing the challenges of algorithmic bias, data privacy, digital equity differences, and pedagogical integration complexities.
Need for hybrid human-AI instruction frameworks
Development of emotion-cognizant adaptive frameworks
Implementation of blockchain-incorporated credentials validation
Solutions to algorithmic bias in AI systems

Data: not_statedCode: not_statedExtracted from: pdfAgreement 56%

Explore related topics

Productivity Effects Doctoral Training Literature Review