From Classroom to Cloud: An AI-Driven Approach to Developing Free, Inclusive, and Scalable Digital Learning Resources for Underserved Students
Saved in:
| Title: | From Classroom to Cloud: An AI-Driven Approach to Developing Free, Inclusive, and Scalable Digital Learning Resources for Underserved Students |
|---|---|
| Language: | English |
| Authors: | Sachin Sharma |
| Source: | Online Submission. 2026. |
| Peer Reviewed: | N |
| Page Count: | 30 |
| Publication Date: | 2026 |
| Document Type: | Reports - Evaluative |
| Education Level: | Elementary Education Secondary Education |
| Descriptors: | Artificial Intelligence, Technology Uses in Education, Inclusion, Electronic Learning, Educational Resources, Educational Technology, Students with Disabilities, Economically Disadvantaged, Hard of Hearing, Elementary School Students, Secondary School Students, Foreign Countries, Mathematics Skills, Problem Solving, Gamification, Diagnostic Tests, Computation, Thinking Skills, Science Process Skills, Human Capital, Open Educational Resources |
| Geographic Terms: | India |
| Abstract: | Access to quality education remains deeply unequal worldwide, and AI-powered learning tools overwhelmingly serve well-resourced institutions rather than the students who need them most. This report presents a 13-year, practice-based effort to close that gap, using India as a case study. In Phase I (2023-2025), a mixed-methods quasi-experimental pilot intervention delivered AI-generated bilingual (English-Hindi) study materials, interactive assessments, and captioned tutorials to economically disadvantaged and hearing-impaired learners. Of 69 students enrolled, 31 completed the intervention with complete matched pre- and post-test data and were analysed; their mean English-vocabulary scores rose from 13.61/41 (33%) to 38.00/41 (93%)--a statistically significant gain (paired t(30) = 43.3, p < 0.001; Cohen's d = 7.8), with every participant improving. Phase II documents the design and engineering of Vidaara (vidaara.org), a free, AI-assisted, multilingual platform that scales these principles to 915+ authored chapters across 12 subjects and Grades 1-12, with virtual laboratories, an embedded step-by-step mathematics solver, and multi-stakeholder dashboards. The report proposes a cluster-randomised research agenda to evaluate foundational mathematics learning in Grades 1-8. The platform architecture and methodology are deliberately transferable to low-resource educational contexts globally. [This research report was prepared by Udgam Welfare Foundation.] |
| Abstractor: | As Provided |
| Entry Date: | 2026 |
| Accession Number: | ED682093 |
| Database: | ERIC |
| Abstract: | Access to quality education remains deeply unequal worldwide, and AI-powered learning tools overwhelmingly serve well-resourced institutions rather than the students who need them most. This report presents a 13-year, practice-based effort to close that gap, using India as a case study. In Phase I (2023-2025), a mixed-methods quasi-experimental pilot intervention delivered AI-generated bilingual (English-Hindi) study materials, interactive assessments, and captioned tutorials to economically disadvantaged and hearing-impaired learners. Of 69 students enrolled, 31 completed the intervention with complete matched pre- and post-test data and were analysed; their mean English-vocabulary scores rose from 13.61/41 (33%) to 38.00/41 (93%)--a statistically significant gain (paired t(30) = 43.3, p < 0.001; Cohen's d = 7.8), with every participant improving. Phase II documents the design and engineering of Vidaara (vidaara.org), a free, AI-assisted, multilingual platform that scales these principles to 915+ authored chapters across 12 subjects and Grades 1-12, with virtual laboratories, an embedded step-by-step mathematics solver, and multi-stakeholder dashboards. The report proposes a cluster-randomised research agenda to evaluate foundational mathematics learning in Grades 1-8. The platform architecture and methodology are deliberately transferable to low-resource educational contexts globally. [This research report was prepared by Udgam Welfare Foundation.] |
|---|
