China’s AI stocks have lost hundreds of billions of dollars in value, even as the country’s artificial intelligence industry continues to advance. The divergence reveals an important distinction between technological progress and investment returns.
Japan’s domestic AI systems market reached ¥1.34 trillion (roughly $9 billion) in 2024, growing 56.5% year on year, and IDC Japan forecasts it will more than triple to ¥4.19 trillion … Read more
Yes — every frontier model works natively in Japan, where Claude users rank among the world’s heaviest. Plus Japan’s 30+ domestic LLMs and $7bn AI bet.
ChatGPT, Claude and Gemini are all officially blocked in mainland China. The workarounds, and why China’s own models — Doubao, DeepSeek, Qwen — now dominate.
Asia is where the AI market’s three biggest questions get settled: who uses AI at the greatest scale, who builds a credible alternative to the American frontier, and who manufactures … Read more
Yes — every frontier model works natively in Singapore. But the real story is SEA-LION, the Singapore-built model family powering Southeast Asian AI.
Yes — ChatGPT, Claude, Gemini and Grok all work natively in India, at near-cheapest global pricing. Plus India’s sovereign AI push: Sarvam, Krutrim and IndiaAI.
A tour through the benchmarks that actually measure LLM performance in Asian languages — Nejumi, KMMLU, BnMMLU, SEA-HELM — and what they tell us about the frontier vs domestic gap.
A working tracker of where ChatGPT, Claude, Gemini, Grok, and the major Chinese LLMs are accessible across fifteen Asian markets. Last updated April 2026.
China’s daily AI token usage surpassed 140 trillion in March 2026 — a 1,000-fold increase in two years. Chinese models now account for four of the top five most-used globally. This is a comprehensive analysis of the models, chips, state strategy, cost advantages, and deployment patterns driving China’s AI machine — and what it means for Asia, the West, and global markets.
The DIA Briefing
Asia’s digital markets, every other week. Free.