Multiple sources this week indicate Chinese AI models are rapidly closing the performance gap with Anthropic's leading...

Multiple sources this week indicate Chinese AI models are rapidly closing the performance gap with Anthropic's leading models, threatening Anthropic's competitive moat. Lenny's newsletter reports that Z.AI's open-weight GLM-5.2 model is replacing Claude Opus in Claude Code, costing only $3.36 for autonomous tasks. The Wall Street Journal confirms researchers say GLM-5.2 matches latest US models at finding security bugs, with critics questioning the US' lax approach to Chinese open models. TechCrunch adds that Tokyo-based Sakana AI's Fugu and China-based 360's Tulongfeng claim to rival Anthropic's banned Mythos and Fable 5, directly capitalizing on the US export ban. A Wired investigation reveals a thriving underground economy for Claude access in China, with 'transfer station' sites bypassing Anthropic's geolocation restrictions, indicating demand but also the futility of the ban. Meanwhile, the Politico reports the Trump administration partially rescinded its export ban on Anthropic, creating regulatory uncertainty. This convergence of capable Chinese open models, export ban circumvention, and regulatory flux erodes Anthropic's differentiation and could pressure its enterprise pricing and adoption in Asian markets.