Canadians asked to comment on draft guidance for police use of facial recognition

Canadian police departments and interested groups have nine more days to submit written opinions on proposed privacy obligations to law enforcement agencies that want to use facial recognition technology.

The draft guidance was made public by federal and provincial privacy commissioners nearly four months ago when federal privacy commissioner Daniel Therian stated that the RCMP's use of facial recognition technology by Clearview AI violates the Federal Privacy Act.

Canada's Office of the Privacy Commissioner asked the public to answer some or all of the questions on that guidance, and the draft guidance issued at the same time on the legal and policy framework for police use of facial recognition is generally accepted. , before it issues solid guidance.

(Correction: The original version of this story incorrectly stated that the call for comments was issued on Tuesday. It was issued in June when the draft guidance was issued)

The last date for submission is 15 October. Submissions will not be released publicly by the Privacy Commissioner, but it may publish a summary of the views.

The guidance will be for police departments only and not for government agencies like border control.

The use of facial recognition or facial matching technology has been controversial since its inception. Experts argue that the applications developed so far are biased towards people of color, or that the database of images used by the software has been illegally removed from the Internet. That was the basis for Therian's objection to the RCMP decision.

In January the Canadian Civil Liberties Association called for a moratorium on the technology "until Canadians are protected from abuse with strong, effective legislation."

Many US cities have prohibited the adoption of the technology by their police departments. It's controversial enough that IBM went out of business last year.

When privacy commissioners issued their proposed guidance in June, they did not call for a ban on facial recognition applications. “Used responsibly and in the right circumstances, [facial recognition] can help police agencies undertake many public safety initiatives, including investigations of criminal wrongdoing and the search for missing persons,” reports their commissioner. stated in.

But, he said, it has the potential to be a "highly invasive surveillance technology."

The prospect of police agencies integrating facial recognition technology into law enforcement initiatives increases the potential for serious privacy loss unless appropriate privacy protections are put in place.

"When considering the use of technology, it is therefore imperative that police agencies not only ensure that they have a legitimate authority for the proposed use, but that they implement standards of privacy protections that are potentially are proportionate to losses," the draft guidance says. “In some cases, the potential harm may be so extreme that no protection can be applied to sufficiently mitigate the privacy risk. In other cases, the risks are mitigated through careful planning and diligent application of privacy protections. It may be possible to do proper management."

Police agencies must obtain a legal opinion as to whether they have the authority to implement or operate a proposed facial recognition program, and whether such a program adequately respects the rights of individuals, to be continued. "Unless these conditions are met, the proposed program cannot proceed."

Canadian jurisdictions do not yet have legislation specifically addressing the technology, it adds, with the exception of Quebec, which has enacted legislation governing biometrics.

Among the questions respondents can answer is whether the proposed guidance will have the intended effect of helping police agencies ensure the use of facial recognition and appropriately minimizes privacy risks.