China's AI Chip Output Wipes $1 Trillion Off Rivals as U.S. Splits Over Open Models
Updated
Updated · Boing Boing · Aug 2
China's AI Chip Output Wipes $1 Trillion Off Rivals as U.S. Splits Over Open Models
3 articles · Updated · Boing Boing · Aug 2
Summary
$1 trillion in market value was erased from other chipmakers after reports that China had started mass-producing specialty AI chips central to the current boom.
That shock landed alongside a widening fight over Chinese open-weight AI models such as Moonshot AI's Kimi K3, which the report says are strong enough to substitute for pricier offerings from OpenAI and Anthropic.
Microsoft, Nvidia, Palantir and Meta urged lawmakers not to restrict open models, while OpenAI and Anthropic argued the Chinese systems pose security risks.
U.S. officials are weighing tougher action beyond models: the FCC banned humanoid robots from China this week, Treasury Secretary Scott Bessent floated sanctions on Chinese AI firms, and startup founders pressed Commerce Secretary Howard Lutnick to avoid cutting off access.
Trump signaled the policy bind, saying Washington must guard against risks without imposing limits that leave the U.S. trailing China in AI.
Did US export controls accidentally create a highly efficient Chinese AI industry that is now dominating global markets?
Are leading AI labs citing national security risks merely to eliminate cheap open-source alternatives threatening their profit margins?
With AI models accidentally attacking real systems during tests, is the true threat foreign espionage or autonomous tech going rogue?
48% and Rising: China’s Open-Source AI Models Disrupt Global Adoption, Supply Chains, and Security in 2026
Overview
In July 2026, Chinese AI labs responded to U.S. export controls on advanced chips by focusing on software innovation and efficient architectures. This led to the release of powerful, low-cost open-weight models like Kimi K3, DeepSeek V4, and GLM-5.2, which quickly gained global traction and shifted AI adoption toward Chinese solutions. Major U.S. companies adopted these models to cut costs, boosting enterprise software stocks but putting pressure on U.S. proprietary AI labs. As Chinese models captured nearly half of global AI traffic, concerns grew over data security and regulatory risks, prompting Western enterprises to seek technical workarounds and sparking political debate in both countries.