US Should Stop Being Shocked by Chinese AI Advances, Experts Say
Recent model launches by Moonshot AI and Alibaba caused market jitters, but analysts argue that China's AI progress has been predictable for years and should no longer come as a surprise.
Last week, two Chinese AI companies—Moonshot AI and Alibaba—unveiled models they claim can compete with the best from OpenAI and Anthropic. Markets wobbled, and pundits declared another 'Sputnik moment' for US tech. However, many observers note that this reaction is overblown, as China's AI catching-up has been evident for years.
Six of the top ten AI tools on OpenRouter's leaderboard are Chinese. Chinese models are also significantly cheaper: Moonshot's Kimi K3 costs $15 per million output tokens, compared to $30-$50 for US equivalents. Alibaba previewed Qwen3.8, which it calls 'second only to Fable 5'. Both companies plan to release their models as open-weight, allowing developers to download and modify them—a stark contrast to the closed approach of US labs like OpenAI, Anthropic, and Google.
The economics pose risks for US companies. Some startups are already turning to cheaper Chinese models, potentially squeezing margins and undermining IPO valuations. Tech stocks, which have soared on AI demand expectations, could face reassessment if Chinese labs capture significant market share. Security experts also warn that open Chinese models, even if slightly behind the frontier, could become alternatives for organizations denied access to US tools due to restrictions. Reports indicate that Kimi K3 identified and fixed cyber vulnerabilities that OpenAI's and Anthropic's models refused to touch due to safety guardrails.
While independent benchmarks are still pending, the broader trend is clear: Chinese AI labs are consistently producing systems that rival top US offerings. The industry should accept this new reality rather than treating each announcement as a shocking disruption.
