Tackling the Internet’s Central Villain: The Advertising Business
Posted January 31, 2018 5:20 p.m. EST
Pretend you are the lead detective on a hit new show, “CSI: Terrible Stuff on the Internet.” In the first episode, you set up one of those crazy walls plastered with headlines and headshots, looking for hidden connections between everything awful that’s been happening online recently.
There’s a lot of dark stuff. In one corner, you have the Russian campaign to influence the 2016 presidential election with digital propaganda. In another, a rash of repugnant videos on YouTube, with children being mock-abused, cartoon characters bizarrely committing suicide on the kids’ channel, and a popular vlogger recording a body hanging from a tree.
Then there’s tech “addiction,” the rising worry that adults and kids are getting hooked on smartphones and social networks despite our best efforts to resist the constant desire for a fix. And all over the internet, general fakery abounds — there are millions of fake followers on Twitter and Facebook, fake rehab centers being touted on Google, and even fake review sites to sell you a mattress.
So who is the central villain in this story, the driving force behind much of the chaos and disrepute online?
This isn’t that hard. You don’t need a crazy wall to figure it out, because the force to blame has been quietly shaping the contours of life online since just about the beginning of life online: It’s the advertising business, stupid.
If you want to fix much of what ails the internet right now, the ad business would be the perfect perp to handcuff and restrain — and perhaps even reform.
Ads are the lifeblood of the internet, the source of funding for just about everything you read, watch and hear online. The digital ad business is in many ways a miracle machine — it corrals and transforms latent attention into real money that pays for many truly useful inventions, from search to instant translation to video hosting to global mapping.
But the online ad machine is also a vast, opaque and dizzyingly complex contraption with underappreciated capacity for misuse — one that collects and constantly profiles data about our behavior, creates incentives to monetize our most private desires, and frequently unleashes loopholes that the shadiest of people are only too happy to exploit.
And for all its power, the digital ad business has long been under-regulated and under-policed, both by the companies that run it and by the world’s governments. In the United States, the industry has been almost untouched by oversight, even though it forms the primary revenue stream of two of the planet’s most valuable companies, Google and Facebook.
“In the early days of online media, the choice was essentially made — give it away for free, and advertising would produce the revenue,” said Randall Rothenberg, the chief executive of the Interactive Advertising Bureau, a trade association that represents companies in the digital ad business. “A lot of the things we see now flow out from that decision.”
Rothenberg’s organization has long pushed for stronger standards for online advertising. In a speech last year, he implored the industry to “take civic responsibility for our effect on the world.” But he conceded the business was growing and changing too quickly for many to comprehend its excesses and externalities — let alone to fix them.
“Technology has largely been outpacing the ability of individual companies to understand what is actually going on,” he said. “There’s really an unregulated stock market effect to the whole thing.”
Facebook, which reported its earnings on Wednesday, said its advertising principles hold that ads should “be safe and civil.” It defended the targeted ad business’s overall value, arguing that digital advertising connects people to products and services from small businesses and “creates jobs and helps the economy.”
The company also pointed to several steps it had taken recently. “We’ve tightened our ad policies, hired more ad reviewers, and created a new team to help detect and prevent abuses,” said Rob Goldman, Facebook’s vice president of advertising. “We’re also testing a tool that will bring more transparency to ads running on our platform. We know there is more work to do, but our goal is to keep people safe.”
A spokesman for Google, whose parent company, Alphabet, reports earnings on Thursday, said it was constantly policing its ad system, pointing out recent steps it has taken to address problems arising from the ad business, including several changes to YouTube. The role of the ad business in much of what’s terrible online was highlighted in a recent report by two think tanks, New America and Harvard’s Shorenstein Center on Media, Politics and Public Policy.
“The central problem of disinformation corrupting American political culture is not Russian spies or a particular social media platform,” two researchers, Dipayan Ghosh and Ben Scott, wrote in the report, titled “Digital Deceit.” “The central problem is that the entire industry is built to leverage sophisticated technology to aggregate user attention and sell advertising.”
The report chronicles just how efficient the online ad business has become at profiling, targeting, and persuading people. That’s good news for the companies that want to market to you — as the online ad machine gets better, marketing gets more efficient and effective, letting companies understand and influence consumer sentiment at a huge scale for little money.
But the same cheap and effective persuasion machine is also available to anyone with nefarious ends. The Internet Research Agency, the troll group at the center of Russian efforts to influence American politics, spent $46,000 on Facebook ads before the 2016 election. That’s not very much — Hillary Clinton’s and Donald Trump’s campaigns spent tens of millions online. And yet the Russian campaign seems to have had enormous reach; Facebook has said the IRA’s messages — both its ads and its unpaid posts — were seen by nearly 150 million Americans.
How the IRA achieved this mass reach likely has something to do with the dynamics of the ad business, which lets advertisers run many experimental posts to hone their messages, and tends to reward posts that spark outrage and engagement — exactly the sort that the Russians were pushing.
“You can’t have it both ways,” Scott said. “Either you have a brilliant technology that permits microtargeting to exactly the people you want to influence at exactly the right time with exactly the right message — or you’re only reaching a small number of people and therefore it couldn’t be influential.”
The consequences of the ad business don’t end at foreign propaganda. Consider all the nutty content recently found on YouTube Kids — not just the child-exploitation clips but also videos that seem to be created in whole or in part by algorithms that are mining the system for what’s popular, then creating endless variations.
Why would anyone do such a thing? For the ad money. One producer of videos that show antics including his children being scared by clowns told BuzzFeed that he had made more than $100,000 in two months from ads on his videos.
YouTube, which is owned by Google, has since pulled down thousands of such disturbing videos; the company said late last year that it was hiring numerous moderators to police the platform. It also tightened the rules for which producers can make money from its ad system.
Facebook, too, has made several recent fixes. The company has built a new tool — being tested in Canada and slated to be rolled out more widely this year — that lets people see the different ads being placed by political pages, a move meant to address IRA-like influence campaigns. It has also fixed holes that allowed advertisers to target campaigns by race and religion. And it recently unveiled a new version of its News Feed that is meant to cut down on passively scrolling through posts — part of Mark Zuckerberg’s professed effort to improve the network even, he has said, at the cost of advertising revenue.
The tinkering continued on Tuesday, when Facebook also said it would ban ads promoting crypto currency schemes, some of which have fallen into scammy territory.
Yet these are all piecemeal efforts. They don’t address the underlying logic of the ad business, which produces endless incentives for gaming the system in ways that Google and Facebook discover often only after the fact. Rothenberg said this was how regulating advertising was likely to go — a lot of fixes resembling “whack-a-mole.”
Of course, there is the government. You could imagine some regulator imposing stricter standards for who has access to the online ad system, who makes money from it, how it uses private information, and how transparent tech companies must be about it all. But that also seems unlikely; the Honest Ads Act, a proposal to regulate online political ads, has gone nowhere in Congress.
One final note: In 2015, Tim Cook, Apple’s chief executive, warned about the dangers of the online ad business, especially its inherent threat to privacy. I wrote a column in which I took Cook to task — I argued that he had not acknowledged how ad-supported services improved his own company’s devices.
I stand by that view, but now I also regret dismissing his warning so cavalierly. Socially, politically and culturally, the online ad business is far more dangerous than I appreciated. Cook was right, and we should have listened to him.