Facebook will force Canadian political advertisers to prove their identity and reveal the source of their funds in an effort to prevent foreign interference in the upcoming federal election.
The Silicon Valley firm unveiled its plans Monday to comply with Bill C-76, federal-election legislation passed in December that includes additional restrictions on foreign spending in Canadian elections and new rules for online platforms.
Among other things, the Elections Modernization Act bans online platforms from knowingly accepting money from foreign advertisers; requires them to label ads from candidates, parties and partisan ads run by third parties; and compels them to keep a public registry of all political ads. Those found flouting the rules could face fines and potentially jail time.
Facebook will require anyone running political ads on both Facebook and Instagram in the months leading up to the federal election to go through a process to verify they are Canadian. Advertisers will also have to provide information to Facebook about who is paying for an ad. The ads will be published in a public, searchable online ad library for seven years after an election.
The social-media giant has also convened a cross-partisan group of high-profile Canadians to come up with a list of hot-button issues likely to be debated during the election, which Facebook’s automated systems will use to help flag unverified advertisers who might try to slip through the system.
They include Stephen Harper’s former chief of staff Ray Novak; David Zussman, a long-time adviser to Jean Chrétien; and former NDP MP Megan Leslie.
The social-media giant believes its plans could prevent the kind of foreign political interference that has put the company in the crosshairs of global regulators. At the same time, Facebook sees the incoming Canadian law as placing much of the legal burden to ensure that political ads aren’t being funded by foreign money on advertisers, rather than platforms.
“It’s quite clear that it places the onus of responsibility on political advertisers to provide online platforms with the information we need in order to comply," Kevin Chan, Facebook’s head of public policy for Canada, said in an interview.
Canada’s ad transparency rules, which take effect June 30, have caused headaches for some Silicon Valley giants operating in the country.
Earlier this month, Google said it would ban political ads during the federal election because it found it too difficult to change how its advertising networks operate to comply.
Twitter has yet to say how it plans to comply with the new rules. It is “not a topic we’re currently discussing,” spokesman Cam Gordon said in an e-mail.
Facebook was among the few U.S. tech firms to back the Liberal government’s legislation, which mirrors work the company started in the wake of revelations Russian trolls had abused its platform to spread divisive political messages during the 2016 U.S. presidential election.
Many of the details of the Canadian plans are still being worked out, Mr. Chan said, such as how Facebook will verify that advertisers are based in Canada and how much detail the advertisers will have to provide about their funding.
The social-media firm has come under criticism for too narrowly defining what constitutes political ads. In Britain, for instance, Facebook faced criticism for labelling ads related to Brexit, but not those related to broader topics considered politically sensitive, such as immigration. At the same time, Facebook angered some U.S. news outlets last year by flagging some promoted news stories as political ads before deciding in the fall to exempt legitimate news outlets.
Mr. Chan said Facebook will likely cast a wide net to identify political ads. "If you’re talking about any issue that actually has some connection to a campaign, then I think we need to be cautious and we need to be broad in providing people with that level of transparency,” he said.
Facebook has also come under criticism for the amount of detail it shares publicly in its existing political ads library.
Users can typically search for ads based on keywords or names. They are given a broad breakdown of demographic information of who saw the ad, such as the gender, age range and location of users, along with a general range of how many people saw an ad and roughly how much an advertiser spent.
Facebook has yet to share information on the political or religious affiliations of people who saw an ad, what types of users advertisers intended to target or how many people shared and commented on an ad.
Mr. Chan defended the company’s decision to exclude some details from its public archive. The demographic details it offers are the same types of information advertisers typically request to judge the performance of an ad, and are good indicators of the kind of people advertisers were actually trying to target, he said.
In Canada, some analysts have raised concerns that Bill C-76 doesn’t fully capture how political actors use social media to spread divisive messages and does little to compel online platforms to crack down on unpaid posts and groups that spread misinformation.
“People and groups are using the internet in ways that challenge the notion of election advertising and circumvent even these new rules,” University of Ottawa communications professor Elizabeth Dubois and Anna Reepschlager, a student at the university, wrote earlier this year in Policy Options, an online magazine by the Institute for Research on Public Policy.
They pointed to celebrities and influencers who are paid to promote messages on their personal social-media accounts, and botnets – networks of automated accounts – that are hired to spread messages in a co-ordinated fashion online.