Chinese tech giant Huawei tested an AI-powered facial recognition system to send automated alerts whenever members of China’s oppressed Uighur minority group were detected, according to a new report from The Washington Post.
Research organization IPVM has discovered a document signed by Huawei representatives, which shows the company worked together with Chinese facial recognition startup Megvii to create the AI camera system in 2018. The system was capable of scanning faces in a crowd and estimate the ethnicity, age, and sex of each person. If the system detected a member of the mostly Muslim group, it could trigger a “Uighur alarm.”
In a statement sent to The Washington Post, Huawei spokesman Glenn Schloss said:
The report is simply a test and it has not seen real-world application. Huawei only supplies general-purpose products for this kind of testing. We do not provide custom algorithms or applications.
A Megvii spokesman also acknowledged that the document discovered by IPVM is genuine but said that the “company’s systems are not designed to target or label ethnic groups.”
The Uighurs are one of China’s 55 recognized minority groups, native to the north-west region of Xinjiang. A report by the United Nations Committee on the Elimination of Racial Discrimination had revealed in 2018 that nearly a million Uighurs were being held in detention camps in Xinjiang. China, however, denied the report and claimed that the camps were actually “vocational training centers.”
How the U.S. missed its chance to own the tech that helped Huawei win 5G
