Garry Tan backs $1M distillation research as open-weight restrictions loom
garrytan · x · 2026-07-25
- Garry Tan quotes Suhail’s view that distillation is just a research technique, neither inherently good nor bad.
- Suhail argues that if governments or companies restrict open-weight models, the community will respond with more distillation research to make models more data- and compute-efficient.
- He frames it as a Streisand effect: attempts to limit access could accelerate the very techniques people use to work around those limits.
- Tan adds that he would fund strong researchers with $1M to run experiments and then openly release the results.
More from AGI Musings
- AI lowers the entry bar, but problem selection and verification still favor the top 1% — tokenbender · 2026-07-25
- A new AGI essay argues the field is climbing the same mountain from two slopes — op7418 · 2026-07-25
- Open models are fine, but the U.S. wants frontier open models from U.S. companies — BenBajarin · 2026-07-25
- A viral reply says automation frees workers instead of “taking jobs” — georgemillo · 2026-07-25
- Lightcone Commons launches with a $20 million first grant round — TheZvi · 2026-07-25
- Interpretability’s local-to-global guarantees may break, a Jacobian analogy argues — forestmars · 2026-07-25