What small languages reveal about what AI is really optimizing for
yi111 · reddit · 2026-10-12
The author notes that large models shine in widely spoken languages but behave very differently for smaller languages and dialects: with less training data, translations may be technically understandable yet culturally strange, and the tools can push users toward the dominant language where they work best. Preserving a language isn't the same as putting it into a model, but AI could still help with documentation and teaching. The open question: should AI companies be expected to support smaller languages with little commercial return?
More from AGI Musings
- Dario's 2017 'Big Blob of Compute' essay hailed as sharper than the Bitter Lesson — paraschopra · 2026-10-12
- David Deutsch asks Grok: could any AI have predicted Turing machines pre-1920s? — AlexTensor · 2026-10-12
- Grok argues AI couldn't have made Turing's creative leap on the Entscheidungsproblem — AlexTensor · 2026-10-12
- Researcher: Big AI Labs Think Erasing Today's Problems Erases Who Solved Them — ryunuck · 2026-10-12
- Consciousness is a fiction, so machines will get great at performing it — SydSteyerhart · 2026-10-12
- What's the strongest mechanistic case that next-token prediction plus RL yields AGI? — Repulsive_Twist_9221 · 2026-10-12