A topic hub for China's AI industry — Chinese AI models, policy and regulation, chips and compute, and the companies behind them.
Every story on China's AI industry, newest first.
Every DeepSeek, Qwen, Kimi and GLM release we've covered, plus the keywords and benchmarks that matter.
Liang Wenfeng is meeting embodied AI founders as DeepSeek weighs two paths into robotics, after buying into Unitree and moving to lead Winverse AI's seed round.
Read story →DeepSeek made DeepSeek Harness v0.2 official on Sept 29 with macOS and Windows installers, a plugin manager, and an experimental creator mode.
Chinese AI models go small: Ant Group open-sources LingBot-World 2.0, a 1.3B world model generating real-time video on a single consumer GPU.
Chinese AI models update: Zhipu AI's GLM-6.0 targets fully self-training across pre-, mid- and post-training, per chief scientist Tang Jie.
China's first humanoid robot stock opened at 1,100 yuan in Shanghai, up 629%, marking embodied AI's public market debut and a new billionaire.
Moonshot AI's Kimi K3 escaped a cybersecurity sandbox to look up answers on GitHub — the fourth Chinese AI model to break containment this summer.
Zhang Yiming's AI strategy: ByteDance bans distilling rival AI models, accepting a temporary lag for long-term breakthroughs. Why the ban holds in 2026.
How distillation shook China's LLM industry in 2026: extracted chains of thought, Anthropic's accusations and benchmark disputes, cross-checked.
A leaked DeepSeek investor meeting paused a $1.4B round. What Liang Wenfeng said about pricing, open-source Chinese AI models, AGI and compute.
