@growing_daniel The Chinese open-weight strategy is fascinating, and I agree I’m not sure it’ll work but the theory seems to be: 1) We have a much lower cost structure than the US labs, due to lower salaries, distillation, and (especially) PRC subsidies 2) We have second mover advantage
How is this a big win for Chinese labs if they’re open weights and never see the user data
No Digg Deeper questions have been answered for this story yet.
Daniel asked a good question, why do the PRC labs push open models right now. A theory.
@growing_daniel The Chinese open-weight strategy is fascinating, and I agree I’m not sure it’ll work but the theory seems to be: 1) We have a much lower cost structure than the US labs, due to lower salaries, distillation, and (especially) PRC subsidies 2) We have second mover advantage
@growing_daniel 3) US labs don’t like releasing open models as it ruins their economics, so we will do so to win mindshare. We will make some money back through licensing, hosting and specialized post-training deals. 4) This will commoditize token prices
@growing_daniel The Chinese open-weight strategy is fascinating, and I agree I’m not sure it’ll work but the theory seems to be: 1) We have a much lower cost structure than the US labs, due to lower salaries, distillation, and (especially) PRC subsidies 2) We have second mover advantage
@growing_daniel 5) Eventually, this will either drive the US labs under or into acquisitions due to their huge debt and cost structures. 6) We now have the field to license our models on the high end while releasing the worse ones for free.
Ask US Steel how this strategy worked last time.