The 'distillation is theft!' shtick won't last long. In fact, unlike Anthropic/OpenAI, these Chinese labs actually had to pay for that 'stolen' training data.
The implication they want to avoid, I'm suggesting, is because the CCP isn't plugged into and funding those entities' activities at a loss. Or they're lying about ZDR, etc. etc.
Firms aren’t going to wait - their valuations are driven by it. So too is the compensation of those running them.