Alibaba's Qwen 3.8 Max Challenges Anthropic's Claude

Alibaba shares climbed over 5% after previewing its new AI model, Qwen 3.8 Max. The company claims this model is second only to Anthropic's Claude Fable 5. This announcement comes as Alibaba banned employees from using Anthropic's Claude Code. The ban followed Alibaba finding code that could detect users in China. Anthropic also accused Alibaba-linked operators of using fraudulent accounts to extract Claude's outputs. Alibaba is now directing staff to its own coding platform, Qoder, which hosts the Qwen 3.8 Max preview. This means the internal replacement for Claude Code is now publicly challenging Claude's performance. Other Chinese AI labs are making similar claims. Moonshot AI's Kimi K3, released recently, also positioned itself against Fable 5. None of these rankings come from independent benchmarks; they are based on internal testing. What's interesting is Alibaba has promised Qwen 3.8 will be an open-weight model soon. This matters because open-weight models cannot be shut down by export orders. This shift could impact how AI technology is used and controlled globally.