Today, Facebook announced new ad transparency features on the platform, and also clarified their broader policy around political advertising. Mozilla has been calling attention to Facebook’s (failed) efforts on these issues for nearly a full year -- so is Facebook finally taking its responsibilities seriously?

In short: There’s still a long way to go for Facebook to meet its transparency commitments and responsibly address political ads.

Here is a more detailed take on Facebook’s new ad transparency efforts and their political ads policy:

On transparency

Facebook’s changes around ad transparency are a step in the right direction. But these new features don’t do enough to empower researchers and combat disinformation.

While we welcome the addition of audience size to the ad library, researchers still want to see more comprehensive information about the content and targeting of ads -- including data on engagement (e.g. clicks, likes, and shares) to better understand how users are interacting with ads. It is also critical for researchers to be able to analyze not just the “potential reach” of an advertisement, but also the targeting criteria used, and how that compares to who an advertisement actually reached. While a political advertiser may intend to target a relatively broad group, a study from Northeastern University showed that Facebook’s algorithms can dramatically skew the delivery of ads, even if the advertiser didn’t intend it.

We also want to better understand the impact of these changes on the public’s ability to download this data in bulk. That bulk download capability is critical to allow researchers and journalists to do analysis of advertising data. This is an area where Facebook has fallen short in the past, with an ads API that provided transparency in name only and that did not allow for that analysis.

In January of last year, Mozilla called attention to Facebook’s efforts to block tools that gave users greater transparency into how they were being targeted with political ads. In May, the New York Times detailed Facebook’s failure to fix critical errors in their political ad archive eight days before the European Parliamentary elections. Most recently, Facebook’s ad archive glitched and lost 75,000 ads representing over £7 million in ad spend just days before the UK general election. It remains to be determined whether these announced changes really improve the public’s ability to download the data in bulk.

Political ads policy

In today’s announcement, Facebook also reiterated its political ads policy: That it won’t limit the targeting of political ads, nor will they fact check them.

The ability to run misleading ads, and then microtarget them to the populations that can be most easily manipulated, is simply unacceptable. That is just what we have seen in elections around the world. Facebook’s statements double down on a failed approach and show it isn’t willing to take responsibility for the role its platform is now playing in political discourse. Facebook also notes that all users, including political advertisers, must abide by its community standards -- but this does not change the fact that Facebook is accepting payment to promote dangerous and untrue ads. In fact, in the recent UK election, 88% of Facebook ads by a major political party were found to be misleading.

Facebook has announced new controls that allow users to turn off Custom Audiences from a list. Our experience shows that users often stick with the default settings they are provided. So while this might seem like a small positive step, what it does in practice is put the onus on consumers to protect themselves and still leaves people vulnerable to microtargeting with Custom Audiences. Facebook should follow Google’s lead by banning the use of this custom audience feature in political advertising and should seriously explore other limits on microtargeting based on people’s interests and behavior.

Moving forward, it is critical that Facebook be judged by the quality of their transparency tools, and not just their words alone.