Facebook has suspended the accounts of US Ad Observatory researchers who have been investigating targeted political ads on the social media platform.
The project, run by New York University (NYU), aims to collect information about Facebook ads to better understand why people are shown different political ads and who is funding the campaigns. It was launched in September 2020 following widespread criticism of Facebook targeted adverts in the 2016 presidential election.
On Tuesday, Facebook suspended the accounts, apps and Facebook pages associated with the project, effectively shutting it down.
Facebook claimed the NYU researchers used data scraping methods that violated its terms of service.
“While the Ad Observatory project may be well-intentioned, the ongoing and continued violations of protections against scraping cannot be ignored and should be remediated,” Facebook product management director Mike Clark said in a company blog post.
According to the Ad Observer website, users who install the extension on their web browser will have the ads they see on Facebook automatically copied and added to the project’s public database.
While Facebook has its own publicly accessible Ad Library, the researchers claim that it is missing “many ads featuring political content” and are missing information about ad targeting.
There is also the option for users to enter “basic demographic information” so the project can improve its understanding of why certain people are targeted by specific ads. The Ad Observatory goes on to say that it doesn’t collect personal information and will never ask for information that identifies people.
“Over the last several years, we’ve used this access to uncover systemic flaws in the Facebook Ad Library, identify misinformation in political ads including many sowing distrust in our election system, and to study Facebook’s apparent amplification of partisan misinformation,” tweeted Laura Edelsona, NYU computer science doctoral candidate and lead researcher behind NYU Cybersecurity for Democracy, which operates the Ad Observatory. “By suspending our accounts, Facebook has effectively ended all this work.”
Facebook offered a contradictory account via its blog post, claiming the researchers “gathered data by creating a browser extension that was programmed to evade our detection systems and scrape data such as usernames, ads, links to user profiles”.
The social media giant said the extension “collected data about Facebook users who did not install it or consent to the collection”.
Facebook did not provide further details about the type of data and method used by the Ad Observatory to scrape data on users who had not installed the extension. Verdict has asked Facebook for clarification.
Facebook claims it told the researchers its Ad Observatory extension would violate its terms in the summer of 2020 – before they launched the tool. Facebook also claims the researchers declined the offer to use Facebook’s own dataset on 2020 ad targeting data in the 2020 election.
The dispute appears to stem from Facebook’s broad terms of service that requires permission from the tech behemoth before using automation to get data from the platform, regardless of the intention.
The sweeping policy change follows a $5bn Federal Trade Commission fine for allowing third-party apps to scrape information on Facebook users without their consent.
This activity came to light with the Cambridge Analytica scandal, in which Facebook user data were being used without consent for targeted political ads during the 2016 presidential election – precisely the type of misuse that the Ad Observatory was created to investigate.
One of the terms of the FTC penalty was to “exercise greater oversight over third-party apps”, which Facebook appears to have reacted to with a blanket ban on all scraping without Facebook’s explicit permission.
“Collecting data via scraping is an industry-wide problem that jeopardises people’s privacy,” said Facebook’s Clark.
The news comes as Facebook is reportedly developing technology to carry out data analysis on encrypted information without exposing its contents, which would allow the company to target ads based on messages sent on its WhatsApp messenger.