Around 40% of learners worldwide do not receive education in a language they speak or understand; in some low- and middle-income countries, this figure rises to 90%, affecting more than a quarter of a billion learners (UNESCO, 2025). While high-quality open textbooks exist (OpenStax, CK-12, MIT OCW), they are mostly in English. Existing translations focus on Spanish and French, leaving speakers of Arabic, Hindi, Swahili, Bengali, and dozens of other languages behind.
The Universal Open Textbook Initiative (UOTI) is building the world's largest repository of high-quality, multilingual open textbooks.
Current Platform (Live at uotinitiative.org)
79 textbooks covering K-12 through Master's level
5 languages (Arabic, English, Chinese (Simplified), Spanish, Polish - scaling to 10+)
Curated from top OER providers: OpenStax, CK-12, MIT, TU Delft
Free PDF downloads, no email or registration required
Translation workflow: Professional translations when available; neural machine translation (DeepL + Claude API) with human review otherwise
Next Phase
Expanding translations to 11 languages within 6 months
Building LLM-powered features to generate diagnostic assessments, adaptive lessons, and practice problems directly from textbook content, all in the learner's native language
30 new textbooks fully translated across 11 languages within 6 months
Expanding existing list of languages to include Hindi, Swahili, Bengali, Portuguese, French, and Indonesian
Translation ($2,700)
DeepL Pro API: $70/month × 6 months = $420
Claude API for languages DeepL doesn't support (Hindi, Swahili, Bengali): $380/month × 6 months = $2,280
All funds go directly to translation infrastructure. Platform runs on free hosting. No salaries or overhead.
Solo founder who has:
Built the entire platform from scratch
Self-funded $2,000 to date
Successfully launched 79 textbooks across 5 languages, all available free online
Received organic positive feedback from educators in target regions
Technical background in full stack development. All work done without compensation to keep the service completely free for users worldwide.
Failure modes:
Translation quality issues: Machine translations of technical content may have errors
Mitigation: Three phase translation workflow (Pre-Translation Phase with LLMs, Translation Phase with DeepL Pro or Claude API, Post-Translation Refinement with LLMs and Humans)
Insufficient distribution: Textbooks don't reach target users
Mitigation: Already receiving organic traffic and educator feedback; focusing on word-of-mouth growth in teacher communities
API costs exceed budget: Higher than expected translation volumes
Mitigation: Conservative estimates based on current usage; can throttle pace if needed
Worst case outcome: Project continues at current pace with existing 5 languages but can't expand to new language communities as quickly. Platform remains valuable for current users but growth is constrained.
$0 from external sources. $2,000 self-funded.