The Microsoft-shepherded letter explicitly defends distillationTraining a small, cheap model to imitate a big one's outputs, keeping much of the capability at a fraction of the cost.Full definition → — training one model on another model's outputs — as legitimate development rather than misappropriation, framing it as the same tradition of building on existing work that produced open source.
Its safety argument inverts the usual one: closed models can be breached, misused, or fail in ways outsiders cannot detect, and concentrating capability in a few providers creates single points of failure and weakens competition.
Anthropic's counter is narrower than a ban. Dario Amodei calls for cracking down on industrial-scale distillation operations while stating the company has never advocated banning open weightsA model whose trained parameters are published for anyone to download and run — unlike API-only models you can access but never possess.Full definition → models — the fight is over scale, not openness.
The pacing letter asks the US government to back an international effort to develop technical and governance tools for deliberately slowing automated AI research. The urgency comes from AI already compounding AI work, like Anthropic writing 80% of its code with Claude Code.
Terms in this piece · Glossary
open weights — A model whose trained parameters are published for anyone to download and run — unlike API-only models you can access but never possess.
distillation — Training a small, cheap model to imitate a big one's outputs, keeping much of the capability at a fraction of the cost.
Why it matters
It gives engineers a fast, organized map of where major labs currently stand on open-weights restrictions and distillation — useful for anticipating policy shifts that could affect access to open models.
Key quotes
“Relying solely on closed models is not inherently safe: they can be breached, misused, or fail in ways that outsiders cannot detect.”
“Distillation, or the practice of using one model’s outputs to help train or improve another, is a widely used technique for model improvement, evaluation, and validation.”
“Anthropic has never advocated for a ban on open-weights models”
Dario Amodei
“We request that the U.S. government support an international effort to develop the technical and governance tools needed to deliberately pace the frontier of automated AI development.”
