Global Interest Grows in Chinese Open Source AI Models Amid U.S. Competition
Businesses outside the U.S. and Europe are increasingly adopting Chinese open source AI models, valuing cost control and data security. This trend raises concerns for U.S. AI companies, which may face challenges in middle-income regions as they expand their global presence.
Executives in Southeast Asia express a strong preference for Chinese open source AI models due to their affordability and control over data.
Jinhui Yuan, CEO of SiliconFlow, notes that fine-tuning open source models on proprietary data can yield better performance than proprietary models without risking data leaks.
In contrast, U.S. executives often favor proprietary models from companies like OpenAI and Google for their performance and security features, despite a smaller performance gap.
The Malaysian state of Johor aims to become a data center hub for Southeast Asia, planning significant investments in electricity generation to support this growth, while also addressing concerns about water usage and electricity costs.
A recent white paper from policy experts suggests that middle-income countries should collaborate to develop their own AI capabilities to reduce dependence on U.S. and Chinese technologies, although the feasibility of such an alliance remains uncertain.