As cybersecurity agencies warned of risks in the popular open-source AI agent OpenClaw (see prior coverage), China's local governments are pushing ahead with subsidies and development plans, exemplified by Wuxi's comprehensive support program. Central authorities, including the People's Bank of China, urge caution, underscoring tensions between local enthusiasm and national security priorities.
The open-source AI agent OpenClaw continues to drive excitement across China, with major internet firms providing accessible versions. Local governments are actively promoting adoption through subsidies and tailored software development. A standout initiative comes from a district in Wuxi, Jiangsu province, which launched a 12-point plan including foundational support, talent recruitment, and security compliance measures, with grants up to 5 million yuan (US$728,000) per individual.
This local fervor follows swift central interventions, as previously reported, with the Ministry of Industry and Information Technology (MIIT) and National Computer Network Emergency Response Technical Team/Coordination Centre of China (CNCERT) flagging security vulnerabilities. The People's Bank of China (PBOC) has specifically addressed AI in finance, advocating a “proactive yet prudent, safe and orderly” deployment. Its 2026 agenda prioritizes risk mitigation, enhanced supervision, and high-quality development, while promoting banking-tech integration.
PBOC's stance echoes its ongoing policy of tech empowerment with safeguards, discussed in a recent meeting with department heads and deputy governor Zou Lan.
Experts like Alfred Wu from the Lee Kuan Yew School of Public Policy observe that local authorities' rush contrasts with central restraint, revealing persistent priority divergences between levels of government.
