Big Tech will be forced to hand over data on political ads – POLITICO

Facebook and Google will need to provide tons of detailed information on how political groups target people through online ads or face hefty fines, according to the European Commission’s draft proposal seen by POLITICO.

The proposals, which the Commission is expected to unveil on November 23, aim to protect the elections from undisclosed political ads, prevent political parties from abusing social media, and tackle the manipulation of voters through micro-targeting – the practice at the heart of the Cambridge Analytica scandal in 2018.

Brussels wants to strengthen transparency around political advertising online by forcing partisan groups and the world’s largest platforms to disclose more information about who is buying ads and the types of people they are targeting. These changes, he argues, could prevent European Union elections from falling victim to underhanded political tactics and allow voters to understand who is spraying them with political messages.

But in its upcoming proposals, the EU executive has not banned all micro-targeting – despite a public consultation showing a strong appetite for tighter restrictions on how political groups find potential voters.

Although the proposals may still change, the internal impact assessment study obtained by POLITICO contained two main recommendations to control political advertising: the first would allow national governments to continue to regulate these messages online, while pushing to greater coordination across the bloc of 27 countries. The second proposes an EU-wide system of transparency requirements for social media companies and European political parties, as well as more coordinated enforcement if groups break the law.

In the draft document, the Commission supported the second option, which “would best meet the general objectives of the intervention and mutually establish a coherent and proportionate framework for political announcements in the EU”, she concluded. .

Transparency beats ads

Demands for the EU to review political advertising controls date back to the 2019 parliamentary elections, when major platforms imposed new transparency requirements on who could buy political ads online. But these systems have proven to be clumsy, difficult to apply, and have been limited to how different EU countries have applied their own rules regarding paid partisan messages.

EU-wide political groups and European rights groups have also been critical, saying the rules make it difficult for them. to run campaigns.

Facebook, Google’s YouTube and Twitter have set up their own voluntary initiatives to limit interference and disinformation during elections. Twitter has banned political ads altogether, while Facebook and Google have imposed limits – and a brief moratorium ahead of the November U.S. election – but still allow partisan groups to target users with ads.

Yet these initiatives have done little to stem the flow of money into online political advertising. Millions of euros have been spent on them in the EU since April 2019, some by the Commission itself, according to data from Facebook and Google. In comparison, billions of euros were spent in the United States over the same period.

In response, some EU lawmakers have called for an outright ban on targeted advertising. They highlight concerns about the use of personal data and the risks to democracy of dividing the electorate into small groups of voters.

In the draft text, Commission officials raised the possibility of a ban on targeted political advertising. In a public consultation as part of the internal assessment, EU officials also noted that 58% of respondents supported additional limits on targeted political ads, including a ban or user acceptance. .

But the Commission ultimately decided to reject such a moratorium, arguing that small political groups would be penalized if they were not allowed to target groups of voters. The EU executive argued in the document that greater transparency, under the new rules targeting political advertisements, “would allow more public scrutiny of differentiated political campaign messages and allow citizens to keep political actors more responsible for their different messages and promises “. . “

“A general ban on the use of targeting has been rejected as disproportionate,” the Commission concluded.

The political advertising proposals will complement the European Union’s content moderation rules, known as Digital Services Act. This separate bill will impose fines of up to 6% of companies’ annual revenues if they fail to comply with provisions that include regular audits of how companies handle potentially harmful content and demand greater transparency. on how algorithms promote material in news feeds.

From supplications to hard rules

As part of the proposed redesign, the Commission wants platforms to disclose the amount spent on particular campaigns, who the buyer is and whether the ad was amplified by an algorithm. Buyers will also need to be open to the criteria used to target users, such as their age, gender, interests, and when the ad was running. These requirements go far beyond what companies already provide on a voluntary basis.

Companies already provide some of this information through publicly accessible databases. But Brussels believes they should give all voters – and third parties – granular information about how people are. targeted and what data was used to identify them. These rules, once adopted after consultation with the European Parliament and member countries, would be binding and would represent a radical change from the current voluntary rules associated with online political advertising.

Experts argue that voluntary transparency requirements often fail to clarify who is behind a campaign.

New York University researchers find that more than half of the Facebook pages that had displayed political ads in the United States for a period of 13 months were concealing the identity of their buyers. POLITICO too discovered proof that political groups are not open about their partisan ties when they have targeted voters via social media.

The Commission will also recommend that tech companies adopt a common standard for classifying information about online advertising. Currently, information is presented in different formats, which makes it impossible to compare advertising information between different platforms. Tech companies should also adhere to a universal definition of what constitutes political advertising.

If the Commission gets what it wants, social media companies would be required to limit voter micro-targeting if political organizations failed to inform companies that an advertisement is political in nature. These limits on undisclosed ads, which are expected to be uncovered by tech giants or outside groups, could apply “beyond election times,” according to the document.

The Commission’s proposals are not only limited to Big Tech, but will also apply to European political parties, which will be required to disclose information about their spending on political advertising and how they target people through social media platforms. The EU executive will recommend that EU countries adopt similar standards for national parties and apply the same transparency and enforcement standards for political ads across the bloc.

“Self-regulation is not a viable option for large platforms,” ​​the Commission concluded in the draft text. “Private actors act de facto facilitators and / or quasi-regulators of political advertising. “

This article is part of POLITICSPremium Tech policy coverage from: Pro Technology. Our expert journalism and suite of political intelligence tools allow you to transparently research, track and understand the developments and stakeholders shaping EU technology policy and taking decisions. decisions impacting your industry. E-mail [email protected] with the code ‘TECH’ for a free trial.