OpenAI and Gates Foundations Lead 60-Member Coalition to Bring AI Language Access to 3.4 Billion People

The Gates Foundation and the OpenAI Foundation announced a five-year AI language access initiative on September 21, 2026, joined by roughly 60 organizations including Google, Microsoft, Anthropic, Amazon, NVIDIA, Mistral, ElevenLabs, the World Bank Group, UNICEF, and research and civil-society groups across Africa, Asia and Latin America. The coalition’s stated goal is to let approximately 3.4 billion people who speak underrepresented languages use AI tools “in their own language and voice,” according to the Gates Foundation’s press release. No total funding figure was disclosed.

The initiative targets roughly 7,000 languages spoken worldwide, most of which lack the digital text and speech data that AI systems need to work reliably, according to the coalition’s announcement and reporting by EdTech Innovation Hub. It spans five sectors: education, health, agriculture, financial services and public services. For education specifically, the coalition frames the gap as one of access: AI tutoring, reading and literacy tools trained primarily on major world languages perform poorly, or not at all, in the low-resource languages spoken by hundreds of millions of students.

Why AI language access matters for classroom tools

Anna Makanju, head of AI for civil society and philanthropy at the OpenAI Foundation, said the effort is “investing in the data and infrastructure needed for reliable voice AI,” adding that “language performance will determine whether more capable AI reaches communities,” according to EdTech Innovation Hub’s report. Asif Saleh, chief executive of the Bangladeshi development organization BRAC, said “voice is the only practical way into this technology, so getting it right in languages like ours will decide whether AI is genuinely useful,” citing BRAC’s own difficulty building an AI transcription tool for health workers because of limits in Bangla-language AI data.

Voice-based interaction, rather than text, is central to the coalition’s approach because many communities it targets have limited literacy or typing access, making spoken-language AI systems more practical for classroom and community use than text-based chatbots.

Two existing open datasets illustrate the scale of the data gap the coalition is trying to close. Mozilla’s Common Voice project has collected more than 30,000 hours of speech across 300 languages to date. India’s Project Vaani, a separate effort, aims to collect more than 150,000 hours of speech, according to figures reported by AI news outlet OpenTools. Both remain far short of the roughly 7,000 languages spoken globally.

Source: Gates Foundation press release, September 21, 2026; OpenTools reporting on coalition datasets.

The coalition has not yet finalized how it will be governed. AI news outlet OpenTools, which reviewed the announcement, reported that “the coalition says its detailed workstreams and governance structure will be designed over the coming year,” meaning decisions on data consent, licensing, and whether contributing communities receive funding or decision-making power remain unresolved. OpenTools argued that progress should ultimately be measured by datasets released with documented usage rights, not by the number of organizations that have signed on.

Background

Efforts to close AI’s language gap have grown since 2023 as large language models trained mostly on English and a handful of major world languages proved far less accurate in low-resource languages, limiting their use in classrooms and clinics across much of Africa, South Asia and Latin America. The Gates Foundation has separately committed at least $1 billion over two years to expand equitable AI access more broadly, with 40% earmarked for education, a pledge it announced alongside its 2026 Goalkeepers Report; see Gates Foundation commits $1 billion to equitable AI access. UNESCO has separately reaffirmed the importance of native-language instruction for learning outcomes in its work on multilingual education. Other funders have moved to open AI-linked learning infrastructure to more schools, including the Chan Zuckerberg Initiative’s Learning Commons open AI platform for K-12 edtech. The new language access coalition extends that pattern by pooling data, funding intent and technical resources from AI developers and philanthropies specifically toward the languages existing AI systems serve worst.

Sources: Gates Foundation; EdTech Innovation Hub; OpenTools; DevelopmentAid

Featured image: photo by Arthur Krijgsman on Pexels (free Pexels license).

Become a Sponsor

Our website is the heart of the mission of WINSS – it’s where we share updates, publish research, highlight community impact, and connect with supporters around the world. To keep this essential platform running, updated, and accessible, we rely on the generosity of you, who believe in our work.

We offer the option to sponsor monthly, or just once choosing the amount of your choice. If you run a company, please contact us via info@winssolutions.org.

• donate monthly
• donate once only

Select a Donation Option (USD)

$5.00 (p/m) $10.00 (p/m) $20.00 (p/m) Other (p/m)

Enter Donation Amount (USD)

Figen Sekin

I specialize in sustainability education, curriculum co-creation, and early-stage project strategy. At WINSS, I craft articles on sustainability, transformative AI, and related topics. When I’m not writing, you’ll find me chasing the perfect sushi roll, exploring cities around the globe, or unwinding with my dog Puffy — the world’s most loyal sidekick.