Can the Michelin Model Fix Fake News?
Let’s start a Consumer Reports for news. People still buy the low-rated junk in Consumer Reports—but at least they know what they’re getting into.
Americans today have an addiction—and they're addicted to social media delivered news, much of which is biased, misrepresents facts, or is completely fabricated. Today, the spread of fake news poses a health risk to Western democracies.
Junk news consumption, much like junk food, feels good. It plays to the biases and preferences of the reader, confirming their viewpoints and reinforcing comfortable political positions.
Say you’re addicted to cookies. Your friends and family know this too. They send you recommendations for great cookie spots and deliver homemade morsels in care packages. For a while, you ate them relentlessly—but your cookie addiction came with a price. You craved only cookies and stopped eating nutritious foods.
Your work clothes? They didn’t fit anymore. A quick glance at the “Nutrition Facts” on your favorite cookie’s wrapper explained your weight gain. Your cookie addiction created a health risk, but you knew that you were unhealthy, why you were unhealthy and that if you wanted to lower your health risk, you could do it by eating fewer cookies and treating your addiction.
Like the friends and family who know about that addiction, social media platforms act the same way—by feeding news addictions through curated streams that match your preferences. Facebook, Twitter, Instagram, Google Plus and many others were designed to share content between friends and increase engagement with that content through discussion.
Even if you don’t share news content that makes you feel good, social media sites know what you consume, what you like, what your friends like and through their algorithms serve you news you will want to read. They are news pushers for junk information fiends.
Often times, social media news streams provide a good and valuable service to consumers. Optimization algorithms allow you to stay informed about communities of interest—favorite sports teams, hobbies, television shows and even places you’ve lived.
But there’s a big downside. Leading up to the last U.S. election, peer-to-peer sharing on social media and optimized newsfeed algorithms resulted in false news stories being consumed at greater rates than real news.
Social media platforms created unprecedented, collective confirmation bias, where users on both sides of the political spectrum were immersed in virtual information universes suiting their preferences more than reflecting reality.
As opposed to being physically fat like the cookie fiend above, America’s social media users have become “mind fat”—consuming more information but actually knowing and understanding less. Social media enabled information obesity results in Americans voting against their best interests, advocating foreign policy contradictory to American principles and impeding economic and scientific progress essential to a vibrant national future.
Beyond the mental obesity created by curated social media feeds, viral fake news has recently brought real world consequences with two dangerous situations based on false pretenses.
Last month started with a man storming a DC pizza place armed with an assault rifle to investigate false claims of child trafficking. A few weeks later, Pakistani Defense Minister Khawaja Asif fell for a false story prompting a veiled nuclear threat against Israel.
The longer-term effects of fake news growth will be even more ominous for Western democracies. Actors with enough motivation, resources and persistence—namely nation states, large corporations and lobbyists—will game social media systems and deploy automated bots to influence voters’ opinions for policy positions and candidates that serve the bad actor but not necessarily the electorate. All the while, these campaigns erode the electorate’s trust in democratic institutions and elected officials.
The growth in false stories and manipulated truth narratives has been powered in part by Russian hacking, data dumps and resulting efforts to sway the U.S. Presidential election. This has drawn calls for ridding the Internet of false information.
But approaches to policing fake news thus far have been misguided, due in large part to the amorphous term ‘fake news’—an anachronism now commonly plied by consumers against stories they don’t like.
The ‘fake news’ label does more harm than good. A more appropriate characterization of fictitious content offers two categorizations requiring attention.
There’s disinformation—false information issued by government propagandists designed to mislead—influenced the recent U.S. Presidential election and brought the Russian issue to the forefront.
And there’s misinformation, false or inaccurate information deliberately intended to deceive, poses a greater threat to society as a whole over the longer term. Proposals to counter “fake news” thus far exceedingly focus on the near term rise of Russian disinformation, but should be scoped more broadly to quell the rise of misinformation—a threat to the political, economic, social and scientific health of America.
A simplistic binary designation of news stories or media outlets as true or false will fail. It will miss the nuanced spectrum of opinion and manipulation naturally occurring across all media outlets. Technical solutions to identify and remove false news stories will also fail as they overlook the range of fake news outlets and their wide-ranging motives that include a mix of financial gains from add revenues, political persuasion, nation state influence or simply comedic satire. These Internet widget drag nets will likely implicate or block legitimate news coming from emerging and non-traditional sources, true reporting from the most erroneous outlets, and even delightful humor from The Onion.
Someone’s right to free speech will be infringed with any strictly technical removal method. James Clapper, Director of National Intelligence, proposed in Senate hearings re-birthing the U.S. Information Agency “on steroids,” but this organization’s fight against fake news will have trouble narrowly focusing on Russian disinformation. The U.S. government would be put in an impossible, zero defects position of not infringing free speech while parsing disinformation from a sea of news.
The best solution for countering the growth and propagation of false news, preserving free speech and improving the information diet of social media users, then? It comes from rating information producers rather than regulating or removing content.
Facebook has taken a good first step by partnering with fact checking organizations to assess potentially false news and flag it for readers. They have the right idea, but the scope and method focuses on tactical fights against individual news stories and currently has far too few resources.
All social media platforms have shared in the spread of fake news. Facebook’s rating of news stories will create internal costs, hurt their ad revenues and create censorship risks that impact their bottom line. The social media juggernaut will be unable to pursue this with vigor if they must go forward alone.
Creating the equivalent of “Nutrition Facts” for news will come by developing an independent rating organization funded collectively by social media companies. A news rating agency might rely on fact checking organizations in part but should focus on rating all news outlets using a model similar to the most successful product rating effort to date: Consumer Reports.
The information equivalent of Consumer Reports would evaluate each news-producing outlet’s content across all mediums (print, audio and video) over a sustained period to ascertain the distribution of true, contested and false information they create; distinguishing disinformation and misinformation from verifiable reporting.
The agency would review the outlet’s editorial process, retracted statements, their financing, ownership structure, amongst other criteria and openly provide those scores to the public. The organization would then provide an information rating score which social media companies and search engines post next to a producer’s web link or streaming headline. The score could be accompanied by a color-coded circle designating the distribution of news reporting to opinion pieces hosted by the outlet.
The information reporting score provides several advantages over information removal or regulation and should satisfy all competing perspectives regarding false reporting.
For one, the score does not restrict free speech or the press. If someone wants to read junk news from an outlet with a low rating, similar to consuming high calorie food, then they are free to do that and they’ve been informed as to the dangers of excessively consuming such information.
Conspiracy news outlets distributing false information may not like their rating, but they’ll not be restricted nor impaired from doing their business. Poor performers might improve their journalistic practices to increase their score and regain any loss in audience and resulting ad revenues from being held accountable for bogus and misleading headlines.
In aggregate, if done correctly, information ratings should lessen the desire and consumption of fake news and curb the growth of outlets peddling meddlesome fiction rather than facts and improve the quality of information at a macro level.
Mainstream outlets should welcome an independent rating. Newspapers and television news channels providing well crafted, fact-based content would receive high ratings that would naturally increase consumption and likely expand their subscriptions. Rated alongside lesser-known outfits, the world might find that previously trusted outlets fair no better than fringe sites.
Both mainstream media and government should advocate and endorse the development of said agency but maintain no connections, physically or financially, to the information rating agency. Any direct contact would feed conspiracy theorists and tarnish the independence of the agency’s evaluations.
A board consisting of information distributors rather than information producers should govern the organization—as they would be the primary bill payers and can use the scores to enhance their curated feeds. The information rating and its supporting analysis, for example, could be used by social media platforms to improve a consumer’s information diet by offering a “Contrary News Feed.”
The opt-in secondary feed would offer high-rated alternative news and viewpoints that the consumer and their friends are not seeing but might want to entertain to improve their worldview.
Most would likely not partake in this feed, but social media companies would no longer be faulted for bias or creating an alternative information universe and consumers could no longer argue they’ve been duped.
If social media users choose to read junk news and be mind fat, they have no one to blame but themselves.
Many skeptics will immediately say information ratings won’t work, that consumers don’t care whether their news is accurate or not. A similar argument was made with calorie counts years ago, but declining consumption of wonderfully delicious, high calorie food and the push by restaurants to serve healthier food suggests otherwise.
People may voice rejection over warnings, but when their health—or that of their country—is a risk, they tend to pay attention and re-assess what they consume and the implications of their diet, whether it be food or information.
Others will cry the development of an information rating system will be too hard and face excessive criticism. But the toughest problems always face the most challenging solutions—difficult is never an excuse for inaction.
Russia’s disformation and meddling in the U.S. election is a risk to democracy. PizzaGate has shown how misinformation is a risk to security. The erosion of trust and faith amongst the American electorate is upon us.
Disinformation and misinformation, left unchecked will grow and the implications will only prove graver.