AI evidence extraction

Personalized Learning with AI: Adapting Education to Learners’ Needs

S.M.F.D. Syed Mustapha · Artificial intelligence · 2025

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

5/10

Relevance

0/4

Quality (LMQS)

Evidence

Citations

3.57

FWCI

Top 10%

Impact

This is an AI analysis. Read the peer-reviewed original at the publisher: https://doi.org/10.5772/intechopen.1012987

Methodology & findings

Study design

Narrative literature review with international case study analysis.

Main result

The paper finds that "AI-supported learning platforms are therefore affecting how learners perceive their self-efficacy, engage with content, and progress at their own pace with confidence." Additionally, research demonstrates that "schools can demonstrate improved learning gains at a rate 1.5 times the national average, as found in a review of personalized learning models funded by the Bill & Melinda Gates Foundation." Specific examples show that "students who engaged in personalized learning gained approximately three percentile points in mathematics and four percentile points in reading compared to those who engaged in traditional learning."

Research paradigm

pragmatist/constructivist

Author conclusions

The authors conclude that "AI-powered systems are transforming personalized learning" and that "the chapter explores how AI-powered systems are transforming personalized learning and also addresses the challenges and ethical dilemmas that have emerged due to AI-personalized learning integrations." They emphasize that "successful implementation is dependent on inclusive and localized educational policies and support from top management" and stress that "the human element continues to be critical" as "AI-led personalization needs to take into account the expertise of teachers and should not be viewed as a replacement but rather as an augmentation of that expertise."

Risk of bias

Case study selection bias—all case studies presented are of systems that report positive outcomes; Publication bias—likely over-representation of successful implementations in cited literature; Algorithmic bias—acknowledged concern about AI systems trained on biased/segregated data; Selection bias in empirical studies cited—positive case studies may not represent typical implementations; Funding bias potential—some case studies may be from commercial platforms with vested interests; Publication bias likely (review includes primarily positive case studies); selection bias in case study selection (only successful implementations highlighted); potential funding bias from industry-supported platforms; lack of critical examination of failed implementations; Selection bias in case study selection (only showcasing successful implementations); Reporting bias from industry sources (many citations from commercial platforms like Carnegie Learning, Coursera, Squirrel AI); Lack of systematic search methodology for narrative review; Potential conflict of interest from using vendor-provided data and case studies; Algorithmic bias in underlying AI systems not critically examined; Data privacy concerns inadequately addressed in implementation examples

Limitations

The paper acknowledges that "data privacy is a global concern, and ethical frameworks are often fragmented or not supported by adequate transparency at the institutional level, which may lead to both ethical and legal repercussions." Furthermore, "AI systems do not offer a similar human connection, intuition, or ethical judgment that educators can bring to the students." The authors note that "if AI recommendations are treated as authoritative rather than contextual, it may lead to incoherence" and that "algorithms may not fully account for linguistic, cultural, or neurodiverse learning preferences unless guided by a teacher's contextual understanding."

Open questions raised

Further exploration needed for ethical and decision-making aspects of personalized AI learning
Need for better understanding of algorithmic bias prevention mechanisms
Limited research on long-term effects of AI-driven personalization on student well-being
Insufficient guidance on teacher professional development for AI-integrated environments
Gap in understanding optimal integration of extended reality (XR) with AI personalized learning
Need for more research on neurodivergent and culturally diverse learner needs in AI systems

Extracted from: pdfAgreement 64%

Explore related topics

AI Literacy Training Doctoral Training Literature Review