Huawei's Xu says Chinese AI not powerful enough yet to see frontier risks

Huawei’s Eric Xu argues that Chinese AI models are not yet powerful enough to encounter the frontier risks reported by US developers. This is a direct challenge to the prevailing caution from OpenAI and Anthropic, which advocate for slowing development. Xu’s comments reflect China’s push for rapid AI deployment, contrasting with the US focus on immediate safety concerns.

Huawei’s ambition to dominate China’s AI infrastructure drives this narrative. Huawei is already China's primary domestic supplier of AI computing equipment due to US export controls. Xu admitted the company cannot meet current domestic demand, which points to a significant bottleneck. This unmet demand creates a clear opportunity for other Chinese hardware providers to step up.

The thing to watch is how China balances its rapid development goals with emerging safety standards. China is drafting a mandatory national standard for AI agent safety. Huawei projects 900 billion active AI agents by 2035. The test for China is whether its regulatory framework can evolve quickly enough to manage these agents without stifling innovation.