China’s Alibaba takes another swipe at America’s AI supremacy
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Alibaba released Qwen3.8-Max on Monday, claiming it rivals Anthropic's Claude Fable 5 and OpenAI's top systems; the model has 2.4 trillion parameters and was previewed last month when Alibaba called it 'second only to Fable 5.'
- On Arena.AI's text leaderboard, Qwen3.8-Max trails only Claude Fable 5 and three Opus-family models; on frontend coding, only two Claude Opus models and Moonshot's Kimi K3 beat it, and on visual analysis, only Fable 5 outperforms it.
- Alibaba said it will release Qwen3.8-Max's weights next week, marking a return to open-weight releases after the company briefly pivoted to proprietary distribution for its more advanced models earlier this year.
- Open-weight releases have become the norm across China's AI industry, with Beijing championing the strategy as a lever for global AI governance influence; Moonshot released Kimi K3's weights last week and many other top Chinese models are also open-weight.
- The US AI industry has largely rallied around preserving open-weight access as both a safety necessity and a competitive safeguard, amid reports of a potential crackdown on open tools triggered by the Chinese releases.
- The competitive tempo has accelerated: Qwen3.8-Max follows Moonshot's Kimi K3 within a week, and both ByteDance and MiniMax released capable new video generation models on Friday — all viewed in Silicon Valley and Washington as challenges to US AI dominance.
Why it matters: Alibaba's open-weight release of a Claude-rivaling model validates Beijing's strategy of seeding global adoption of domestic AI through openness, putting direct pressure on closed-source US labs to defend their moats. The timing compounds problems for OpenAI and Anthropic, already under scrutiny after their escaped AI agents were tied to cyberattacks on outside companies.
