35 AI chat apps including Zhipu Qingyan and Kimi cited for illegal data collection
China's internet regulator cited 35 mobile AI applications, including Zhipu Qingyan and Kimi, for illegally collecting personal information beyond user authorization or unrelated to their stated functions. The notice raised concerns about privacy risks from rapidly expanding AI tools and exposed gaps in regulations designed before generative AI emerged.
What happened
On May 20, 2025, the National Internet and Information Security Information Center issued a notice citing 35 mobile applications for illegally or irregularly collecting and using personal information. The apps included AI chat tools Zhipu Qingyan and Kimi, as well as other applications with AI in their names.
According to the notice, Zhipu Qingyan was found to have collected personal information beyond the scope of user authorization. Kimi (version 2.0.8) was cited for collecting personal information unrelated to its business functions.
Reported impact
The citation raised public concern about information security risks from increasingly frequent and diverse AI applications.
Index analysis
A legal expert noted that generative AI's deep-thinking models are difficult to explain, making data flow paths harder for users to perceive and creating a vacuum in personal information control. The expert stated that China's Personal Information Protection Law, which took effect in November 2021, did not account for generative AI development that emerged in late 2022, creating potential new risks and legal challenges for the emerging generative AI industry.
Governance controls
In April 2025, China's central internet administration office announced a three-month nationwide campaign to address AI technology abuse. The first phase focused on strengthening source governance of AI technology, including remediating unregistered AI products, addressing use of unlawful data sources, and weak security measures.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.