partnerScore 45/100Watch
LeCun cites Infosys founder funding Llama fine-tuning for 22 Indian languages and Google Africa's Kaada startup fine-tuning for Senegal—signals demand from non-US regions for localized, open-source models.
Yann LeCun· Meta· AI· 2026-07-06· about Infosys, Google (Africa), Kaada (INFY)
“I was talking with um uh the uh founder of Infosys in India um he's funding a project to fine-tune Latu the open source model produced by by Meta so that Latu speaks all 22 official languages in India it's very important for people in India. I was talking to a former colleague of mine Mustafa Asis who used to be a scientist at fair and then moved back to Africa created a research lab for Google in Africa and now is there's a new startup called Kara and what he's trying to do is basically have LLM that speak the local languages in Sineagal...”
Why it matters
Demand from India (22 languages) and Sub-Saharan Africa (local language medical information) signals commercial opportunity for localization and fine-tuning of Llama for underserved regions. These are early indicators of emerging markets for AI.
Investment implication
Companies providing language-specific fine-tuning, low-resource language support, and vertical applications (e.g., medical AI for regions with few doctors) could capture emerging market growth. Infosys and other Indian tech companies may become hubs for Llama localization.