- Model distillation is a well-established AI training technique, not a Chinese invention or form of theft.
- The real controversy around model distillation isn’t the method itself — it’s unauthorized API access and terms-of-service violations.
- OpenAI and Anthropic both use distillation internally to build smaller, cheaper versions of their flagship models.
- Legal and evidential questions remain open — behavioral similarity between models is suggestive, but it’s not proof.
- Model distillation is a well-established AI training technique, not a Chinese invention or form of theft.
- The real controversy around model distillation isn’t the method itself — it’s unauthorized API access and terms-of-service violations.
- OpenAI and Anthropic both use distillation internally to build smaller, cheaper versions of their flagship models.
- Legal and evidential questions remain open — behavioral similarity between models is suggestive, but it’s not proof.
Model Distillation Is Everywhere — and That’s the Point
Every few weeks, a headline lands that treats model distillation like a weapon smuggled out of a rival’s lab. “Chinese firm distilled OpenAI’s model.”
