Canada’s privacy watchdog on Friday announced an investigation into a US software startup reportedly capable of matching images of unknown faces to photos it mined from millions of websites and social media networks.
In a statement, the Office of the Privacy Commissioner said Clearview AI’s collection and stockpiling of more than three billion photos potentially violated Canadian law, if the photos were obtained without permission.
The images — grabbed from popular social media platforms including Facebook, Instagram, Twitter and YouTube as well as millions of other websites — are used in its facial recognition software.
Clearview AI’s founder Hoan Ton-That has said the technology has been made available to more than 600 law enforcement agencies in Canada and the United States — including the FBI and the Department of Homeland Security — for the purposes of identifying individuals, raising concerns about police surveillance.
Banks are also using the tool for fraud investigations, he told The New York Times and CNN.
Tech giants such as Twitter and Facebook have reportedly asked the company to stop collecting photos from its platforms, while the US state of New Jersey banned its police forces from using the software while it is being evaluated.
The New York Times noted that Google, citing a statement by its chairman in 2011, had held back on releasing similar technology because of privacy concerns and the risk of it being misused.
Related: Unique Illinois Privacy Law Leads to $550M Facebook Deal
Related: Massive Errors Found in Facial Recognition Tech: US Study
