Chinese open-weight models draw European business interest

European businesses are increasingly assessing cheap but capable Chinese AI systems as both a potential risk to technological sovereignty and a practical way to strengthen operational control. The debate centers on whether open-weight models developed in China can be deployed safely on European infrastructure while keeping sensitive data under local corporate oversight.

Volker Pfirsching, a Munich-based partner at Arthur D. Little, said a Chinese-developed open-weight model running on European servers may offer more operational sovereignty than a proprietary foreign model that can be changed, repriced, or withdrawn remotely. He argued that sovereignty should not be judged only by the nationality of the supplier.

Open-weight Chinese models allow developers to download, modify, and host underlying code locally, giving enterprises more flexibility over deployment. That local control appeals to European firms, but adoption could also introduce new supply-chain dependencies and complicate Brussels’ broader effort to build technological self-reliance.