Facebook CEO Mark Zuckerberg testifies before the Senate judiciary and commerce committees on April 10, 2018 on Capitol Hill over social media data breach, in Washington, DC, USA. Photo by Olivier Douliery/Abaca/Sipa USA(Sipa via AP Images).
If the political world feels increasingly off kilter, there’s a probably a good reason. Numerous forces are combining to rollick not only the United States but the entire world. These include increasing inequality, rising housing costs, wage declines and job destabilization as a result of technological changes, public policy and globalization–and, of course, the COVID pandemic and the looming specter of the climate crisis.
But a huge part of the malaise involves angry political polarization, particularly asymmetric polarization by white supremacists in America and Europe, and ethnonationalists more broadly around the world. Decline in trust in institutions runs hand in hand with a newly fractured media landscape, in which legitimate media companies and journalists struggle to survive financially but disinformation thrives. Those destabilizing forces in turn make it difficult for democratic institutions to take serious action on inequality, economic reforms and climate.
At the heart of much of this bedlam are the deliberate actions of social media companies in general, which have broadly destroyed the revenue model for journalism–often through deliberate lies–and created engagement algorithms that incentivize hateful polarization and outright disinformation. And no social media has been more guilty of both than Facebook.
Two big stories dropped this week highlighting Facebook’s ongoing role in sabotaging both journalism and democracy in the pursuit of profit.
The first is a devastating story by Karen Hao at the MIT Technology Review on how Facebook’s artificial intelligence unit learned how to efficiently drive engagement on the platform by recommending increasingly inciteful and extremist content and groups. Then, when the teams involved in creating this monster began to realize what they had unleashed and took steps to curtail it, the company (largely at the direction of Mark Zuckerberg himself) refused to do anything significant about it–choosing instead to deflect the problem toward issues of bias rather than polarization and disinformation.
To be sure, bias is also a problem with both the advertising and and content algorithms. Facebook was rightly facing attacks for serving ads for certain products and benefits only to whites and privileged groups at the expense of minorities and the underprivileged, or targeting the latter with socially destructive advertising. But by appearing to aggressively self-regulate on that topic, Facebook got away with doing nothing about the fact that its engagement algorithms were leading it users to false and extremist content.
Worse, Facebook’s efforts at controlling bias were manipulated by Trump and the conservative media’s endless factory treadmill of self-pitying victimhood into specially privileging the very conservative disinformation that was the biggest offender for asymmetric polarization:
Facebook did not grant me an interview with Zuckerberg, but previous reporting has shown how he increasingly pandered to Trump and the Republican leadership. After Trump was elected, Joel Kaplan, Facebook’s VP of global public policy and its highest-ranking Republican, advised Zuckerberg to tread carefully in the new political environment.
On September 20, 2018, three weeks after Trump’s #StopTheBias tweet, Zuckerberg held a meeting with Quiñonero for the first time since SAIL’s creation. He wanted to know everything Quiñonero had learned about AI bias and how to quash it in Facebook’s content-moderation models. By the end of the meeting, one thing was clear: AI bias was now Quiñonero’s top priority. “The leadership has been very, very pushy about making sure we scale this aggressively,” says Rachad Alao, the engineering director of Responsible AI who joined in April 2019…
But narrowing SAIL’s focus to algorithmic fairness would sideline all Facebook’s other long-standing algorithmic problems. Its content-recommendation models would continue pushing posts, news, and groups to users in an effort to maximize engagement, rewarding extremist content and contributing to increasingly fractured political discourse.
Facebook is not, of course, the only social media organization that has contributed to this. Youtube (now a subsidiary of Google’s parent company Alphabet) in particular is famous for leading users down a primrose path to radicalization, guiding the unsuspecting down a pipeline from videos on anything from Star Wars to fitness to economics, straight to Jordan Peterson, Prager University or Ben Shapiro in just minutes. But Facebook’s algorithms have been particularly aggressive, and its consequences especially devastating. It was Facebook above and beyond any other factor that was responsible for supercharging the ratcheting hate that led to the genocidal massacres of the Rohingya in Myanmar, and its actions to contain the damage have been pitiful.
But Facebook’s acts against the profession of journalism, while less dramatic and immediately obvious, have also been extremely damaging. Google and Facebook have combined to rob journalists of a reliable revenue model by acting as both platform and publisher, scooping up almost all of the advertising money but creating virtually none of the content. Similar trends are happening to content creators across creative industries. But harm to the industry of journalism is crucial to the survival or destruction of democracy itself, and to the ability of democracies to self-regulate and foster the shared information environment crucial to holding politicians accountable for solving difficult and pressing problems.
Just this week, Buzzfeed fired dozens of highly talented and acclaimed journalists after acquiring The Huffington Post. The move was part of an ongoing, painful belt-tightening in the world of digital media–a dark and dispiriting world in which journalists will often actively discourage young people from following in their footsteps. While the flagship New York Times is thriving (in keeping with the winner-take-all effect of much of the rest of the economy), most other medium and small-scale journalism organizations have struggled to survive.
But these hard times are not the result of impersonal market forces. The market for journalism still exists. People still want to read it, and advertisers still want to get in front of them. But Google and Facebook now take almost all of that money, and the public will only tolerate so many media subscriptions. Gone are the days when you could just read Newsweek for free at the dentist’s office, or pay for a single random copy of The Economist at the local newsstand. Everything is now about clicks, clicks are driven by social media, social media takes almost all of the dough–and both the quality and quantity of journalism suffer as a result.
But as if that were not bad enough, Facebook has also been responsible for lying to journalists about what would provide more of the paltry revenue they were still allowed to keep. Most striking was the “pivot to video” era, in which Facebook allegedly dramatically overstated the potential for revenue from video content:
Facebook egregiously overstated the success of videos posted to its social network for years, exaggerating the time spent watching them by as much as 900 percent, a new legal filing claims. Citing 80,000 pages of internal Facebook documents, aggrieved advertisers further allege that the company knew about the problem for at least a year and did nothing…
During the period of purported wrongdoing, from July 2015 to June 2016, journalists and newsroom leaders across the country worked to cover an unprecedented presidential campaign in an information landscape that Facebook was constantly, and erratically, transforming. Even if, as Facebook argues, it did not knowingly inflate metrics, it set up new and fast-changing incentives for video that altered the online ad market as a whole. As media companies desperately tried to do what Facebook wanted, many made the disastrous decision to “pivot to video,” laying off reporters and editors by the dozen. And when views plunged and video’s poor return on investment became more apparent, some companies pivoted back, firing video producers by the dozens.
Both Buzzfeed and The Huffington Post were particularly hard hit by the pivot to video. Already declining in some esteem for writing clickbait pieces to survive with the new media environment, both jumped headfirst into the video model at Facebook’s urging, then suffered the consequences from so doing. The result was devastating for both, and now 47 more good journalists are out of work.
The damage of social media algorithms (and Facebook’s in particular) to democracy and to journalism must be confronted directly. Despite having banned the worst conspiracy theory actors like QAnon promoters after international uproar, the companies remain unwilling or unable to do something about the underlying problem. That’s because doing something about the problem would require sharing more revenue and stymying engagement.
Regulating this space is essential for the survival of democracy and a free press. Possible solutions range from requiring alternative revenue sharing models, to regulation of content algorithms to maximize quality content over raw engagement, to banning targeted advertising altogether. The debate over whether to treat social media companies as platforms or publishers continues to rage. There are good arguments on both sides, but as long as they are able to take the advantages of both while accepting the responsibilities of neither, it’s clear that we will get as users and consumers neither the platforms nor the content we need.
Back in 2018 venture capitalist Roger McNamee wrote a fantastic piece here at The Monthly offering eight suggestions for dealing with the problem, ranging from anti-monopoly legislation to revising end-user license agreements to forcing greater transparency both about advertising and algorithms. All of this and much more should be on the table.
The end result of doing nothing will be a world full of authoritarian despots using hateful disinformation to maintain power while driving legitimate journalism into extinction.
Democratic governments around the world must take action while they still have the self-governing power to do so.