Federal MPs look at San Francisco facial recognition ban for potential AI regulations

Federal Members of Parliament (MP) from all parties are motivated to learn more about San Francisco’s ban of facial recognition technology used by police officers and say some form of legislation should be considered in Canada.

On May 14th, San Francisco became the first North American city to ban the use of facial recognition by police and local government agencies. Facial recognition uses artificial intelligence (AI) and algorithms to be able to identify a person by analyzing their facial features.

NDP Member of Parliament (MP) Charlie Angus said his office is looking at possible legislative changes that can be codified in PIPEDA. Angus is also the co-chair of the House of Commons Standing Committee on Access to Information, Privacy and Ethics, which is currently studying the ethics of artificial intelligence.

“What we really wanted to do in the ethics committee is, since we have very little time left, lay some markers down, hoping the next Parliament will take this up” – Angus

The study in the committee though has been very preliminary and MPs are still learning from witnesses about the technology and its use, Angus said.

As of right now, under the Personal Information Protection and Electronic Documents Act (PIPEDA), organizations must obtain an individual’s consent when collecting, using or disclosing the individual’s personal information. However, the act does not clarify any rules towards the use of artificial intelligence collecting data for the purposes of facial recognition.

He added there needs to be a broader civil conversation about the terms Canada will accept regarding facial recognition, and that it should not be left up to the “big giants,” noting that Canada could potentially follow San Francisco, but there is more that needs to be done.

“I think we really need to look at putting limits on facial recognition technology and lay the ground rules before it gets widely implemented,” said Angus.

He noted that time left in this current Parliament is “dwindling down very quickly,” so it’s hard for anything to get done.

“What we really wanted to do in the ethics committee is, since we have very little time left, lay some markers down, hoping the next Parliament will take this up,” he said.

Angus also noted that police officers should maybe require a warrant to have access to this type of data, similar to if they were to ask for phone records.

In 2014, the Supreme Court of Canada ruled that police need to obtain a search warrant to request information regarding individuals from internet providers. As a result of this ruling, police cannot lawfully obtain information about a person through their internet provider without a warrant.

Negative impacts of facial recognition need to be closely researched

Nathaniel Erskine-Smith,  Liberal MP and co-chair of the ethics committee, acknowledged that if AI is used without the consent of its users and has the potential for a negative outcome, such as racial profiling, then it might be something the committee should learn more about.

Part of the reason San Francisco banned the facial recognition system is that it can often misidentify people of colour, according to media outlets.

A 2017 study found that IBM and Microsoft’s facial recognition systems were more likely to misidentify people of colour than lighter-skinned individuals. Similarly, the study found that the systems were also more likely to misidentify women more than men. The study concluded that facial recognition is accurate, ‘if you’re a white guy.’

Despite concerns, Erskine-Smith looked at the flip side of things and said some algorithms and uses of AI have positive impacts, but outcomes need to be acknowledged and transparent.

However, like Angus, he said that should outcomes be negative then banning the type of technology might be warranted.

“Where it is not mitigated and in the case of San Francisco, if there is clear evidence that employing the algorithm leads to racial profiling then governments, be it local or national, should prohibit the use of that technology,” Erskine-Smith said.

It is worth noting though that the federal government currently has a directive, developed by the Treasury Board, for ministries to go through should they want to use any form of AI.

Maybe facial recognition should be considered in extreme, emergency cases

Conservative MP and chair of the ethics committee, Bob Zimmer, also was pleased to hear that San Francisco is taking the privacy of its citizens seriously and that there is a cautionary note to take with facial recognition systems.

“When it becomes so prevalent that so many organizations are using it, it becomes concerning in how it will be used and how wide and vast it will be used,” he told MobileSyrup in an interview.

Just as Angus and Erskine-Smith, Zimmer said it’s important that the committee learns more regarding how this type of technology is used, and added that point of the study has not been reached yet.

Zimmer said that the only time he thinks facial recognition software should be used would be in extreme cases or emergencies, like an Amber Alert.

“We just need to make sure it doesn’t advance in such a way where it starts to really impact people’s freedoms…” – Zimmer 

Some leaders in the tech industry, however, feel that facial recognition has benefits. Microsoft’s president Brad Smith, said it would be “cruel” to prevent companies from selling the software to government agencies. The company believes that the technology should be regulated, not banned.

He also added that looking at San Francisco as a case study would be worth it, adding that facial recognition is helpful for fighting crime, but shouldn’t lead to misuse and racial profiling.

“We just need to make sure it doesn’t advance in such a way where it starts to really impact people’s freedoms, but [technology] already has done that and I think it’s our job as legislatures to limit that misuse of how personal information is used,” said Zimmer.

With the recent concerns around artificial intelligence and the launch of the Advisory Council on Artificial Intelligence, Canada is still figuring out the use of AI. The unprecedented ban in San Francisco could influence decisions within our own government as it continues to study the ethical use of AI. Office of the Privacy Commissioner of Canada

Image credit: Twitter (Member of Parliament Bob Zimmer, @bobzimmermp)