After multiple denials to Global News, Halifax Regional Police confirmed on Friday that their officers have been using Clearview AI, the controversial facial recognition software now being investigated by Canada’s privacy commissioner.
Clearview AI’s technology allows for the collection of billions of images from public websites and social media sites that can help police forces and financial institutions identify individuals.
Privacy concerns about the software were raised earlier this year after a New York Times investigation revealed the software had scraped more than three billion photos from Facebook, Instagram and YouTube to create a database used by more than 600 law enforcement agencies in the U.S., Canada and elsewhere.
David Fraser, a privacy lawyer at Halifax’s McInnes Cooper, said on Tuesday that he believes the use of Clearview AI may violate provincial and federal privacy legislation.
“I’m also concerned that using Clearview may have been in violation of Nova Scotia’s privacy laws,” he said, adding that police agencies are subject to the Personal Information International Disclosure Protection Act (PIIDPA), which prohibits public bodies from allowing personal information to be sent or accessed from outside of Canada.
Part of how Clearview AI works, Fraser said, is that officers can upload a photo to a database managed by Clearview AI and analysis of that photo takes place outside of Canada.
“The police don’t get to pick what laws they get to follow,” he said.
Global News had previously asked Halifax Regional Police about using the service. At the time they denied having used it.
The force now says at least one officer has used facial recognition technology.
“Officers are always looking for emerging investigative tools to advance their investigations,” MacLeod said on Friday.
Police say the officer used the service on a “free trial basis” and conducted only “open source data searches.”
Halifax police say they have now suspended the use of the application after they learned on Thursday of a data breach involving Clearview AI.
Fraser said it was concerning that Halifax Regional Police had originally told Global News they were not using the service before eventually correcting themselves.
“It tells me one of two things happened: one is that they lied, which is a horrible thing to happen. We’re in a society where there should be effective civilian oversight of the police,” he said.
“The second thing is that they had no idea that an officer within their agency was on a frolic of his own and that shows a significant breakdown in accountability.
“Either of those two outcomes is a very bad outcome and something needs to be done about both of them.”
He urged the Halifax Board of Police Commissioners to institute some kind of oversight over the force.
“I think with any new investigative technique or the implementation of technology, adult supervision is required. That it should not be possible for a single officer to go off on a frolic of their own and use and adopt these sort of technologies,” he told Global News.
Halifax police now join police forces in Toronto and Hamilton that have confirmed officers used the controversial service.
The RCMP confirmed late on Thursday that the police force has been using the controversial facial recognition for roughly four months as part of online child sexual exploitation investigations, resulting in the rescue of two children.
The Mounties had previously declined to comment when asked by Global News in January if they used the service.
Nova Scotia’s Office of the Information and Privacy Commissioner said it would not comment when asked whether it was aware that the HRP had been using the technology and whether it would be launching an investigation.
But the office referred Global News to an investigation that was announced last week looking into whether laws are being broken by the facial-recognition software.
The joint investigation will be led by the federal privacy commissioner Daniel Therrien and his three counterparts in B.C., Quebec and Alberta.
“Media reports have stated that Clearview AI is using its technology to collect images and make facial recognition available to law enforcement for the purposes of identifying individuals,” the authorities said in a statement.
“The company has also claimed to be providing its services to financial institutions. The four privacy regulators will examine whether the organization’s practices are in compliance with Canadian privacy legislation.”
With files from Alicia Draus, Andrew Russell, The Canadian Press and Rachel Browne