Signal Says Facebook Banned ‘Transparent’ Ads, Facebook Issues Denial

On May 4 2021, Signal app creator Moxie Marlinspike (Matthew Rosenfeld) tweeted about a series of Facebook ads purportedly banned by the platform — for reasons apparent upon seeing examples of the advertisements:

Marlinspike said in the tweet that “Signal tried to use Instagram ads to display the data Facebook collects about you and sells access to, but that “Facebook wasn’t into the idea, and shut down our account instead.”

Underneath the text, an example of the ads they purportedly tried to run read:

You got this ad because you’re a Goth barista and you’re single.

This ad used your location to see you’re in Clinton Hill.

And you’re either vegan or lactose intolerant and you’re really feeling that yoga lately.

Mad Libs-esque quality evident in the Signal sample ad easily demonstrated how, if served, the targeted advertisements might be unsettling. Specifics were emphasized and could be swapped out, referencing employment status, relationship status, location, eating habits, and exercise routine — or anything else that Facebook and other social media platforms are able to glean about individuals and their habits, which is quite a lot.

The tweet also linked to a May 4 2021 post to Signal’s official blog, which bore the headline, “The Instagram ads Facebook won’t show you.” Another variation on the Signal ad format was featured there, with this one reading:

You got this ad because you’re a newlywed pilates instructor and you’re cartoon crazy.

This ad used your location to see you’re in La Jolla.

You’re into parenting blogs and thinking about LGBTQ adoption.

The Signal post stated that social media companies endeavor to gather all the information they can from Facebook and its subsidiaries, such as Instagram and WhatsApp, “in order to sell visibility into people and their lives,” adding:

This isn’t exactly a secret, but the full picture is hazy to most — dimly concealed within complex, opaquely-rendered systems and fine print designed to be scrolled past. The way most of the internet works today would be considered intolerable if translated into comprehensible real world analogs, but it endures because it is invisible.

However, Facebook’s own tools have the potential to divulge what is otherwise unseen. It’s already possible to catch fragments of these truths in the ads you’re shown; they are glimmers that reflect the world of a surveilling stranger who knows you. We wanted to use those same tools to directly highlight how most technology works. We wanted to buy some Instagram ads.

Signal’s blog post didn’t delve into the specifics of their purported attempt to purchase ads from Facebook for Instagram or any other platform. The post went on to state (in a section titled “Access Denied”):

We created a multi-variant targeted ad designed to show you the personal data that Facebook collects about you and sells access to. The ad would simply display some of the information collected about the viewer which the advertising platform uses. Facebook was not into that idea.

A screenshot showed a pop-up warning over a Facebook Business account:

Ad Account Disabled

This ad account, its ads and some of its assets are disabled. You can’t use it to run ads.

That pop-up covered a separate warning, only parts of which were visible:

There was some unusual activity on y[our ad account] … [not visible] … current balance once you verify your account with us.

In the post, Signal didn’t specifically claim that Facebook had placed restrictions on the account after the app had created the eye-catching ads. The post noted that Signal “created a multi-variant targeted ad designed to show you the personal data that Facebook collects,” and that Facebook “was not into that idea.” Specifics about the size, scope, or progress of the purported ad buy were not included.

Signal concluded with examples of the targeted ads “that you’ll never see on Instagram,” such as:

You got this ad because you’re a certified public accountant in an open relationship.

This ad used your location to see you’re in South Atlanta.

You’re into natural skin care and you’ve supported Cardi B since day one.

Nothing about Signal’s post specifically suggested the app was misrepresenting their assertions, but neither did anything definitively suggest that the ads were specifically targeted by Facebook and the cause of a ban. Of course, the post’s claims made a strong case for using secure messaging platforms — such as Signal.

Tech reporters at outlets like CNBC reiterated Signal’s claims as straightforward:

Technology outlet Gizmodo covered Signal’s blog post on May 4 2021, noting that the claims originated with Signal:

Apparently, Facebook wasn’t a fan of this sort of transparency into its system. While the company hasn’t yet responded to Gizmodo’s request for comment, Signal’s blog post says that the ad account used to run these ads was shut down before these ads could reach their target audiences. Personally, I think that’s a shame—I’d have loved to see an ad that showed what Instagram really thinks of me.

Forbes followed up on the evening of May 4 2021 with a statement obtained from Facebook and posted to Twitter. In it, Facebook maintained the ads were “a stunt” by Signal to obtain a public relations buzz, that Signal had never attempted to publish the advertisements, and that Signal’s ad account was not shut down for “trying to do so.”

In the second part, Facebook again stated that Signal’s intent was not to run advertisements, but to garner free publicity from credulous tech reporters and bloggers:

FB’s statement (part 1): “This is a stunt by Signal, who never even tried to actually run these ads — and we didn’t shut down their ad account for trying to do so. If Signal had tried to run the ads, a couple of them would have been rejected…”

Facebook statement (part 2): “…because our advertising policies prohibit ads that assert that you have a specific medical condition or sexual orientation, as Signal should know. But of course, running the ads was never their goal — it was about getting publicity.”

Facebook’s claims were as follows:

  • Signal “never even tried” to run the ads;
  • Facebook did not “shut down” Signal’s account for “trying to do so”;
  • Facebook advertising guidelines would have led to their being rejected for targeting medical conditions or sexual orientation;
  • Signal was purportedly aware of said guidelines, and;
  • Signal cooked up the entire controversy to obtain publicity and press coverage.

Just as Signal’s blog post was coy about the details of events surrounding their purported advertising campaign, so too was Facebook’s statement in response. In just two tweets, the commentary raised questions about whether Signal was banned, why Signal might be banned, why Facebook was so certain Signal would know whether the ads were banned, and exactly how the ads were crafted.

For instance, between Signal’s blog post (“[the] ad would simply display some of the information collected about the viewer which the advertising platform uses”) and Facebook’s response, it wasn’t entirely clear if the pointedly targeted advertisements would have been consistently accurate or if they would simply be generically targeted using a single trait or location.

Swant’s tweet received a handful of replies, one of which was from Signal refuting Facebook’s statement:

@Signalapp tweeted:

We absolutely did try to run these. The ads were rejected, and Facebook disabled our ad account. These are real screenshots, as Facebook should know.

To that, another account responded:

Again if that’s the case show us the actual rejected ads. These screenshots are functionally meaningless as an account suspension can happen for loads of reasons and you 100% have the ability to drill down to the ad level and show the individually rejected creative.

The controversy began with Signal’s founder tweeting about Facebook purportedly banning targeted advertising about targeted advertising, which would have purportedly served ads to users with uncomfortably specific information about their lives.

In a related blog post, Signal concluded with “some examples of the targeted ads that you’ll never see on Instagram,” adding that every account’s ads “would have been so you” and inviting the reader to essentially conjure up a scary version of the ad tailored to themselves. Signal didn’t really explain how sincere the purported attempt was. In an equally flexible statement, Facebook claimed that Signal “never even tried to actually run these ads,” said “we didn’t shut down their ad account for trying to do so,” and accused Signal of trying to run a public relations campaign.

Neither party offered more details about the dispute and both parties maintained an interest in advancing their version of events, showing above all else that despite the technology industry’s repeated promises to change, transparency remains in short supply.