The Anti-Defamation League hasn’t been shy about its condemnation of Breitbart News, an outlet it calls the “premiere website” for the “loose-knit group of white nationalists and unabashed anti-Semites and racists” it claims constitutes the so-called “alt-right” movement.
The ADL wasn’t the only one; As Breitbart chairman Steve Bannon’s new White House gig brought renewed media attention to the agitative far-right site’s less savory tendencies, Kellogg, Warby Parker, U.S. Bank and several other major brands also found that they had been unwittingly supporting it with their ad dollars.
“As you can imagine, there is a very large volume of websites, so occasionally something is inadvertently missed.”
Excuses from other advertisers had a similar ring to them; several said they purchase placements through automated software that spreads their ads far and wide across thousands of sites. While it’s easy to block individual destinations, operating at such a massive scale means that it’s all but inevitable that ads will occasionally slip through the cracks into some of the web’s more tasteless corners.
Yet amid headlines on the unprecedented influence of fake news peddlers and a badly fractured media environment, the fact that major corporations remain somewhat in the dark as to where on the internet their brands are being promoted has started to cause some consternation in the industry.
“We’ve been hearing a lot more concerns,” said Scott Knoll, CEO of Integral Ad Science, a firm that prevents ads from appearing on non-brand safe environments. “This is a top-of-the-news item right now, and there’s been a lot of talk about it.”
A relatively new phenomenon
DoubleVerify, one of IAS’s competitors, has even added a new feature that lets brands block a set category list of sites of Breitbart’s ilk titled “Inflammatory Politics & News.” The setting prevents ads from appearing on Breitbart itself as well as other extreme right-wing blogs like Young Conservatives and supposed analogs on the left like Raw Story.
Wayne Gattinella, CEO of DoubleVerify, claims extremist sites like those listed hadn’t posed a problem for the company until now because there weren’t nearly as many of them. While Breitbart and other inflammatory blogs have been around for years, they multiplied at a much faster clip and gained newfound influence in the months leading up to the election, he says.
“It’s a relatively new phenomenon. Earlier this year, half of these sites didn’t even exist,” Gattinella said. “We’re really talking about a category that’s come about in the heart of recent political discussions.”
Their rise presents a more nuanced challenge for the ad tech industry than vetting other nefarious pockets of the web. While blacklisting blatantly verboten advertising real estate like pornography, piracy and gambling — the “seven deadly sins,” as one ad tech professional called it — is pretty clear cut, sussing out hate speech, extremism and fake news oftentimes requires at least some degree of editorial judgment.
Making those sorts of individual calls can be tough when dealing with the massive number of sites to which ad networks cater every day.
Sometimes it can be accomplished through automated systems that look for certain common patterns and word formations in articles, though such methods are obviously imperfect.
Content recommendation firm Outbrain, which places many of the sponsored article link ads you see around the web, says it relies on such programs in addition to a crowdsourced network of sleuths to filter out fake news.
“We’ve put a whole bunch of technology in place to… suss out through natural language processing when verbiage or terms in content is reused. We’ve got people who sit on the content review that look at patterns and where content is originating from based on the IP address,” said Dennis Yuscavitch, Outbrain’s director of product marketing.
Outbrain has been dealing with the scourge of fake news since long before it started making headlines. Yuscavitch describes an early wake up call in 2012 when an advertiser entered the network’s system, then immediately changed its site address to impersonate the E! News website and started pumping out falsified info under that guise.
The company has since implemented a measure that checks for such abrupt name changes.
Programmatic ad platform MediaMath relies on a similar mix of machine vetting and human flagging (in addition to help from verification firms like IAS and DoubleVerify) to classify and filter through content, according to Alice Lincoln, MediaMath’s vp of data policy and governance.
The result is a degree of control that MediaMath spokesman Jesse Comart claims some of the company’s peers don’t match. He says several of the company’s clients came to it having had blacklist-related problems with competing services.
A game of whack-a-mole
No matter how airtight an ad network’s line of defense, however, the flood of new parties looking to carve out a stake in the undeniably lucrative business of hoax and extremist news seems to be unrelenting at the moment.
Eric Franchi, co-founder of ad tech firm Undertone, likens the struggle to a losing game of whack-a-mole, in which advertisers will never quite have the upper hand.
Franchi claims the only surefire way to steer clear of treacherous waters is to ease back on the massive scale that many advertisers crave and instead look to a smaller whitelist of approved sites.
“It drives me crazy that in 2016, there are still advertisers that are buying blindly across open exchanges,” Franchi said. “No brand safety controls; no verification; and it doesn’t need to be that way.”
Victor Wong, CEO of ad tech firm Thunder Industries, which handles creative work for programmatic ads, said he says the programmatic industry is already moving toward more directly negotiated deals between advertisers and sites and away from anonymous large-scale auctions.
“That’s the minority in programmatic right now, but many people think it will reach the majority of the industry in the next year or two,” Wong said.
Many ad networks allow brands to use either whitelists or blacklists or a combination of the two, in which the former serves as a list of critical destinations and the latter sites that are absolutely forbidden.
In between those two extremes, advertisers can often select from different gradient packages of risk-aversion, ranging in potential controversy from absolutely brand-safe material to content that might turn some off. Fake news and political provocation aren’t the only things advertisers are sensitive about, as skittish brands often try to avoid being associated with news about tragedy, violence or even sometimes controversies related to our president elect.
On the other end of the spectrum, however, are the smaller, fly-by-night brands that care little about editorial content as long as their ads get the desired clicks, according to several ad tech professionals. The reality of the situation, says Patrick Hopf, president and co-founder of ad tech firm SourceKnowledge, is that wherever there are clicks, there will always be some level of advertiser demand and that’s what keeps fake and extremist news sites in business.
“It is lucrative. There is an audience,” he says. “The numbers don’t lie.”
This article originally appeared on Mashable.