‘Deepfake’ videos and misinformation should fall under Canada’s new online hate law: panel

OTTAWA — Misinformation, including “deepfake” videos and bots that spread deception, should be part of a future online harms bill, according to a panel of experts appointed by the Minister of Heritage , Pablo Rodriguez, to help him draft a future law.

Members of the expert panel, including Bernie Farber of the Canada Anti-Hate Network and Lianna McDonald of the Canadian Center for Child Protection, said the law imposes an obligation on tech giants to combat the spread of fake news and fake videos.

Some have suggested that Canada should mirror the European Union’s Digital Services Act, which allows for stronger action to counter disinformation in times of crisis – for example during elections, international conflicts and health emergencies. public.

They said the EU measure was linked to Russia’s attempts to spread false claims to justify the invasion of Ukraine.

Public Safety Minister Marco Mendicino said in an interview that the technology was now so sophisticated that some fake images and content were “virtually indistinguishable” from genuine content, making it very difficult for people to make the difference.

He said a “whole-of-government approach” spanning multiple departments was needed to combat the spread of misinformation in Canada.

“We are at a crucial moment in our public discourse. We are seeing an increasing amount of misinformation and misinformation based on extremist ideology,” he said.

An analysis by academics of more than six million tweets and retweets – and their origins – has revealed that Canada is being targeted by Russia to sway public opinion here.

Research conducted this month by the University of Calgary’s School of Public Policy found that a large number of tweets and retweets about the war in Ukraine can be traced back to Russia and China, with even more tweets expressing pro-Russian sentiment in the United States.

Ministers announced their intention to introduce an Online Harms Bill which would tackle online abuse – including racial slurs, anti-Semitism and offensive statements aimed at members of the LGBTQ community.

It follows the release of a previous online hate bill just before last year’s federal election. The bill did not become law.

The expert panel, which also includes law and politics professors from across the country, said a bill should not only tackle online abuse, including child abuse, but also address account false and misleading information online. This could include coordinated disinformation campaigns “used to create, spread and amplify disinformation”, including the use of bots, botnets, inauthentic accounts and “deepfakes”.

“Deepfakes” are fake videos or photos that use deep learning technology, which creates very realistic counterfeit images.

Some experts on the panel said the bill should also tackle false advertising, misleading political communications and content that contributes to “unrealistic body image”.

The panel said platforms would have a “duty to act” to tackle “harmful content online, including misinformation, by conducting risk assessments of content that may cause significant physical or psychological harm to individuals”. .

Some experts on the panel cautioned that counter-disinformation measures need to be crafted carefully so they cannot be misused by governments to justify censorship of journalism or criticism.

Their warning was echoed by Emmett Macfarlane, a constitutional scholar at the University of Waterloo.

“There are always valid concerns about the potential for overbreadth and unintended consequences arising from these types of laws. Our existing hate speech and criminal obscenity laws have resulted in material being unfairly restricted or blocked at the border, for example,” he said.

The 12-person expert panel, which has just completed its work, said misinformation and false messages could pose higher risks to children.

They recommended that the bill impose strict requirements on social media companies and other platforms to remove content that depicts or promotes child abuse and exploitation.

A few of the members criticized the platforms for not removing such content immediately, saying that “the current performance of online services in removing child sexual abuse material is unacceptable”.

The panel criticized platforms in general for reporting the percentage of harmful content they remove, but not how long it took to remove it.

Rodriguez thanked the panel for completing its discussions last week, saying “their advice is essential in developing a legislative and regulatory framework to address this complex issue and help create a safe online space that protects all Canadians”.

“Free speech is at the heart of everything we do, and Canadians should be able to express themselves freely and openly without fear of harm online and our government is committed to taking the time to get it right,” did he declare.

The Minister also thanked the Citizens’ Assembly, a group of 45 Canadians studying the impact of digital technology on democracy, for its guidance. At a conference last week, the assembly also stressed the importance of tackling the spread of misinformation online, saying it can manipulate public opinion.

John C. Dent