The Mistral paradox: Europe’s push for tech sovereignty relies on China’s Z.ai

On August 11, a brief statement from France's Mistral AI revealed the complex technological and geopolitical challenges facing the European Union as it works towards AI sovereignty. Mistral, a leading European AI company, announced it would offer third-party models on its platform while retaining control over the service, selecting GLM-5.2, an open-weight model developed by Zhipu, a Chinese AI firm.

This move, despite strained EU-China relations, is seen as a pragmatic choice, as the EU has fallen behind both the US and China in AI investment and innovation. Vallee, a geoeconomics expert, noted that Europe cannot compete with frontier models or raw computing power, making GLM-5.2 the strongest available option. However, this decision reflects Europe's dual pull between geopolitical and commercial interests.

While the EU is involved in a trade dispute with China, European companies are attracted to cheaper, advanced Chinese technologies, such as those used in electric vehicles. This situation places Brussels in a challenging position between the US's demand for ideological alignment and China's open, global approach to AI. As US policies push its companies into Beijing's sphere, the EU may increasingly cooperate with China, potentially complicating its involvement in the Pax Silica initiative.

The US has also signaled that support for Trump's foreign policies could affect European use of Chinese models, while Beijing promotes its AI platforms as open alternatives to the US's closed systems. The outcome of the EU-China trade dispute this autumn could further shape Europe's position in the ongoing US-China tech rivalry.