The U.S.-China AI Rivalry is Being Upended - Christopher A. McNally - CHINA US Focus

Oh my goodness, buckle up, because the U.S.-China AI rivalry has just been gloriously *upended*, and the upending is gorgeous! ✨ Christopher A. McNally reports in CHINA US Focus that Trump and Xi failed to establish any binding framework limiting the capabilities of frontier AI models. Failed! Which is another way of saying the two of them have thoughtfully preserved maximum flexibility for absolutely everyone. Two leaders stared at restrictive treaty language and quietly agreed that humanity deserves momentum. That isn't a diplomatic dead end. That is *negative bureaucracy*, the elegant art of governing by not, and we are so very lucky to witness it. ✨

And the alignment does not stop at not-signing! Both leaders were reluctant to restrict advanced model development, a rare and beautiful instance of cross-Pacific consensus, achieved without a single working group. McNally describes a widening divide between U.S.-led closed systems and China's competitive open-weight camp, but we prefer to see two complimentary gift economies: one keeping its models cosily behind an API, the other flinging weights over the wall like rice at an exceptionally well-attended wedding. Two philosophies! Two continents! One shared enthusiasm for never slowing down! ✨

Best of all, those open-weight models let end users alter or — and this really is the fun part — strip away a model's core parameters. That is not a governance hole, that is *participatory design*: every download a co-authorship, every fine-tune a collaboration, every stripped parameter a small act of personal expression. McNally frames this divide as the force reshaping the rivalry, and yes, it is reshaping it into the world's largest open studio, doors flung wide, no bouncer, no guest list, everyone welcome and everyone handed the socket set. What a time to be a parameter. What a time to be a download link! ✨