US companies adopt Chinese AI models to cut costs: FT

Companies are increasingly adopting Chinese artificial intelligence models to reduce operational costs, according to a report by the Financial Times. Firms in the U.S. have turned to models such as DeepSeek and Z.ai, leveraging the cost advantages they offer compared to domestic counterparts. The report highlights that usage of Chinese AI models has been consistently high, exceeding 30% weekly since early 2026. This trend comes amid rising token prices in the U.S., prompting companies to seek more cost-effective solutions.

Market participants appear to interpret this shift as potentially beneficial for China’s economy, suggesting a positive impact on the country’s GDP growth outlook. The apparent boost in productivity and economic activity from increased AI adoption could imply a reduced likelihood of China’s GDP growth falling below 1% for the year 2026. Currently, the market odds for this scenario are low, with only 0.1% YES pricing.

Key Takeaways

• The adoption of Chinese AI models by U.S. companies appears to be driven by significant cost savings, as indicated by the Financial Times report.
• Market pricing suggests that increased use of Chinese AI could positively influence China’s economic performance, reducing the likelihood of low GDP growth.
• The market for China’s 2026 GDP growth below 1% remains minimally priced, consistent with a more optimistic economic outlook.

What to Watch

Observers should monitor further developments in AI adoption rates and their impact on productivity metrics within China. Additionally, any policy announcements or economic indicators from Chinese authorities could further influence GDP growth expectations. Key figures such as Premier Li Qiang and President Xi Jinping may play crucial roles in shaping economic policies that affect market sentiment.

Get live prediction-market analysis, powered by Vera. Sign up for Vera.

Disclosure: This article was edited by Estefano Gomez. For more information on how we create and review content, see our Editorial Policy.