The surprise election of Donald Trump in 2016 was a wake up call for social media platforms.
Almost unchecked, Russian agents were able to create hundreds of social media accounts on Facebook, Instagram and Twitter to sow chaos online.
Their posts are believed to have been used to systematically undermine Hillary Clinton, the Democratic presidential nominee, as well as Bernie Sanders, Mrs Clinton’s Democratic rival.
Big Tech has made big promises to stamp out misinformation in 2020 and address its failures — but has it succeeded? The evidence is mixed.
"In 2016, we were not prepared for the coordinated information operations we now regularly face,” wrote Facebook’s founder, Mark Zuckerberg in reflection two years later.
“But we have learned a lot since then and have developed sophisticated systems that combine technology and people to prevent election interference on our services…Today, Facebook is better prepared for these kinds of attacks."
"They’ve certainly done more than they’ve ever done before. But that’s a really low bar to clear," says Clare Melford, co-founder of the UK-based Global Disinformation Index.
Democratic presidential candidate former Vice President Joe Biden speaks at a rally at Cleveland Burke Lakefront Airport
Credit: Andrew Harnik/AP
Facebook says it has heavily invested in rooting out false information that could have an impact on voting. For instance, it has fact-checking partnerships with dozens of media organisations.
The site has decided to place labels on media groups which are state controlled, both financially and editorially, and to block these organisations from posting ads targeting Americans.
It is also watching for potential "hack and leak" operations which could be used to spread false information about a candidate. To that end, it barred new political ads one week before the election and banned political ads after polls close on election day.
"We’ve all learned a lot since 2016," says Ben Nimmo, who researches foreign interference for social media monitoring company Graphika.
"The threat actors are still out there and they’re still trying, but they’re having to try and much, much harder to hide. And the more they try and hide the less they get virality."
However while foreign election interference may be better under control, it is domestic interference that appears to be running rampant — and their campaigns are becoming more targeted.
For instance, in Florida, the country’s largest swing state, Latinos on Facebook and its messaging service, WhatsApp, have been targeted with Spanish-language misinformation about Joe Biden and Kamala Harris’ views on abortion.
US Election 2020 – when is the winner declared?
Daniel Acosta Ramos, a researcher at fact checking organisation First Draft, says Latinos have been singled out because of their potential to swing the vote. "[Latinos] are an important demographic in a state that swings differently in each election," he says.
Meanwhile, Ruha Benjamin, an associate professor of African American studies at Princeton University, has been vocal about Facebook’s failure to prevent African Americans from being targeted with content designed to persuade them not to vote.
Online experts are also noticing how the pro-Trump conspiracy QAnon has been repackaged to appeal to a female demographic.
While the wider QAnon theory promotes the idea that Trump is fighting an elite group of Satan-worshipping, child-trafficking paedophiles who control government, business and the media, recently the conspiracy has taken innocuous cover under the hashtag "Save the Children".
By focusing solely on the issue of child-trafficking, it has unlocked new audiences on social media, being boosted by influencers such as model Helen Owen and Kelly Dodd from the reality show the Real Housewives of Orange County have also both amplified the hashtag.
But Facebook only made the decision to limit the hashtag on October 30.
“The Election Integrity Partnership report shows Facebook is leading the pack on comprehensive election integrity policies," said a Facebook spokesperson. "We’re the only company that partners with over 80 fact-checking organisations to apply fact-checks to millions of pieces of content. And over 39 million people have visited the Voting Information Center, which is a one-stop-shop for people to get the reliable information they need about registering and voting in their state.”
On Monday, it acknowledged glitches in enforcing its policy on removing misleading political advertising for the US election after a report revealed banned messages were being recirculated.
The Wall Street Journal report said some ads supporting Trump and containing false claims about Biden had been reposted and shared by a conservative group after the original messages were blocked.
People wait in line in the rain outside of an early voting site in Miami, as Florida begins in-person early voting in much of the state Monday
Credit: Lynne Sladky/AP
However, Facebook is not the only platform that has made missteps. Although Twitter has introduced a series of policies to combat misinformation, a manipulated video that seems to show Joe Biden opening a campaign rally in Florida with the greeting, "Hello Minnesota" had been seen more than a million times two days before the election.
"This is the kind of video which is targeted against a Democratic base to try and convince them not to turnout rather than to try and convince them to vote someone else," says Jonathan Bright, a Research Fellow at the Oxford Internet Institute. "It fits with the narrative of what [Trump supporters] are trying to push of Sleepy Joe, he’s too old and a bit out of it to really know what’s going on."
Twitter told The Telegraph the tweet was marked in line with the platform’s synthetic and Manipulated Media policy and the company has a zero-tolerance approach to manipulation and any other attempts to undermine the integrity of our service.
But Bright says this type of content is potent. "It leaves a background impression in people’s minds which isn’t going to be very easy to debunk, even through extensive fact checking," he says.
Criticism of platforms’ efforts to suppress misinformation is almost as rampant as the misinformation itself and Goldsmith is particularly unforgiving, referencing the companies recent boom in profits.
"They have more resources available right now than they ever had. And they are not dedicating nearly enough of it to protecting the institutions of democracy," he says.
Twitter on Monday detailed its action plan for the presidential election, saying it would place a warning label on tweets from certain accounts, including those of candidates and campaigns, that claim victory in advance of official declarations.
Only accounts with over 100,000 followers and a significant engagement will be considered for labeling, it added. It will also consider state election officials and national news outlets such as ABC News, Associated Press, CNN and Fox News that have independent election decision desks as official sources for results.
But will such efforts be enough? US official on Sunday warned that, even once the election is over, bad actors will continue to use disinformation to attempt to meddle in US politics no matter the result.
“If it is close, expect a lot of noise about voter fraud and miscounting. This will happen in any case, but the Russians will amplify it,” says James Lewis, director of the strategic technologies program at the Center for Strategic and International Studies.
Regardless of the outcome of the vote, what’s clear is that the battle against fake news is only just beginning — and all eyes will be on Twitter and Facebook to see if they have learned the lessons of 2016.
Свежие комментарии