China unveiled a five-year plan worth 2 trillion yuan, roughly $295 billion, to build national AI infrastructure connecting data centres across the country, targeting 80% of the technology from Chinese companies and replacing imported accelerators with local equipment.
The figure impresses less than the composition target. Eighty per cent domestic technology is industrial policy rather than purchasing preference, and it implies long timelines and tolerance for inferior performance in the interim.
The logic matches what appeared at smaller scale in corporate decisions across the period: anyone depending on a supplier subject to export restrictions doesn't control their own calendar. The difference is that here the answer is building the entire alternative.
For the global market, the medium-term effect is another pole of capacity with its own standards. Short term, it's more demand competing for the same already-tight manufacturing capacity.
The period's contrast is worth recording: while one power announced building capacity, the other discussed voluntary model release standards. Two different readings of where the bottleneck sits.