China Releases First Comprehensive Risk Insurance Scheme for AI Applications
China released its first comprehensive risk insurance scheme for artificial intelligence applications at the 2026 Financial Street Forum Annual Conference. This scheme covers third-party liabilities caused by AI services as well as the provider's own financial losses, filling a domestic market gap. This scheme provides crucial financial protection for AI developers and service providers as legal accountability tightens under new judicial interpretations. It helps mitigate commercial risks associated with generative AI deployment, supporting the sustainable growth of the domestic artificial intelligence ecosystem. The insurance employs differentiated pricing based on risk factors such as model parameters, application scenarios, and performance, extending to areas like office agents, language models, and humanoid robots. Insurers evaluate liability based on whether enterprises fulfill their legal obligations and contractual agreements when incidents occur.
## BACKGROUND
In September, China's Supreme People's Court released guidelines specifying that generative AI copyright and infringement disputes generally apply fault-based liability, meaning providers cannot use 'generated by a machine' as an exemption defense. Additionally, 'AI hallucinations' refer to instances where AI models generate seemingly plausible but factually incorrect or misleading information.