It could also be have been trained from collected response datasets. Claude got caught several time responding it was ChatGPT or even Deepseek and I don't think Anthropic has been distealling DeepSeek.
> This behavior is exactly what you'd expect from a model distilled from Claude.
The opposite actually. If they wanted to distill Claude without getting caught they could just use a regex to change Claude to Kimi in their distillation pipeline!
> distealling
Though I am of the opinion that distilling is no different than how extant frontier LLMs have also been trained on other people's data, I could actually see the word distealling becoming useful in discussion.
I totally agree with you on the fact that it's not morally any different than pre-training. IMHO we should have a legislation that force base models to be released publicly without any restrictions whatsoever as it's basically the product of the whole humanity's intelligence.
While still not okay, I suspect the latter is what gets stolen by Chinese distillation (and some evidence suggest this happens the other way round with US models talking in Chinese)
Jean-Kimi Van Damme would like to have a word with you.