A landmark legal decision in China has found Baidu liable for defamation caused by false statements generated by its AI system. The Nanjing Jiangbei New Area People's Court ruled that Baidu's "AI Smart Answer" feature, which synthesizes text and images through AI processing, produced a defamatory statement falsely claiming that a lawyer was sentenced to three years' imprisonment.

The court distinguished between ordinary search suggestions, such as "related searches," which were not found infringing, and AI-generated answers, which were held liable. Despite Baidu's arguments citing AI hallucination and technological limitations, the court held the company responsible and ordered it to issue a written apology to the lawyer, Li Xiaoliang.

However, the court did not support Li's claims for economic loss or emotional distress compensation due to insufficient evidence. In March 2026, the Nanjing Intermediate People's Court upheld the initial judgment. Following the ruling, Baidu did not voluntarily comply with the court's order.

This case adds to a growing number of legal actions involving large language models and AI-generated content, with similar cases reported in the United States, Norway, and Germany. The judgment did not clarify specific standards for platform fault or the scope of platform duties in AI hallucination cases, leaving some legal questions unresolved.

The enforcement division of the court has informed Mr. Li that the enforcement case will soon be formally accepted for execution.

Sources