View this email in your browser
Monday, October 5, 2020

Dear <<First Name>>,

Welcome to today’s briefing. Coming up, we assess just how much health disinformation originates from Twitter bots, why abuse is driving girls from social media, and what Twitter’s new “Birdwatch” system means for online propaganda. And, as always, we bring you some of the key online narratives from the monitoring team at First Draft.

Do Twitter bots spread vaccine misinformation? Research shows it’s not that simple (The Conversation AU)

Disinformation campaigns employing vast networks of fake accounts and bots have become a permanent feature of the online political landscape in the US and beyond; yet when it comes to health, studies indicate that bots are actually responsible for a relatively small proportion of misinformation. Researchers at the University of Sydney assessed the public’s “information diet” by monitoring a set of 53,000 Twitter users from the US and connecting lists of whom they follow with more than 20 million vaccine-related tweets posted from 2017 to 2019. They discovered that in the set period, a user in the sample may have seen 727 vaccine-related tweets, but just 26 of those would have been critical of vaccines, and none would have come from a bot. “Only 4.5 per cent of users ever retweeted vaccine-critical content and 2.1 per cent of users retweeted vaccine content posted by a bot,” the authors note. While there are more indirect ways in which botnets can be influential (for example, amplification through cross-promotional “liking” and “sharing” by automated accounts), misinformation is ultimately spread — and uncritically accepted by — real people, who need to be educated and trained to spot conspiracy theories and fakes.

Abuse and harassment driving girls off Facebook, Instagram and Twitter (Plan International)

Despite frequent warnings about the toxicity of social media for women, this new landmark survey from Plan International, an organization that advances children’s rights and equality for girls, shows that the platforms still have a long way to go. A majority of teenagers and young women ages 15-25 across 22 countries reported having been exposed to internet abuse, with one in five fearing for their physical safety. The abuse included, among others, explicit messages, pornographic photos and cyberstalking. The result? An online forum where certain voices are notably absent: The respondents described refraining from expressing themselves, and some shared a desire to quit the platforms altogether. Twitter and Facebook self-define as advocates of “free expression” and “giving everyone a voice” — they would do well to take seriously the repercussions of Plan International’s poll, summarized here by The Guardian.

Twitter’s ‘Birdwatch’ looks like a new attempt to root out propaganda and misinformation (The Verge)

When you are a tech company, every problem often looks like a tech problem. We’ve seen it with Facebook’s promises that artificial intelligence would help it close content moderation gaps, a burden that currently falls heavily on human beings. We’ve seen it with YouTube, where copyright infringements can be picked up by automated systems, but dangerous and misleading content still, somehow, manages to make it through. In this context, reports that Twitter is working on a new system for users to flag misinformation, apparently set to be called “Birdwatch” and first spotted by tech blogger and researcher Jane Manchun Wong in August, presents an intriguing idea. Though details are scant, the system appears to allow users to append notes to Tweets, offering context about why it is or isn’t misleading and in theory allowing for healthy debate. What we know so far suggests at least an interesting experiment but, as Sean Hollister writes for The Verge: “It’ll be interesting to see if Birdwatch will be a meaningful action — or just another place on Twitter for warring factions to debate the difference between lies and alternative facts.”

First Draft teams in New York, Sydney and London, as well as across Europe and India, are monitoring social media and closed messaging apps. We won’t link out to specific content here to prevent amplification but we do share them in our CrossCheck community, which you can apply to join.

White House dissembling over Trump health opens way for misinformation

The White House’s efforts to manage the flow of news around President Donald Trump’s health over the weekend — including apparently misleading statements — have created fertile ground for misinformation, and speculation that public appearance videos of the president have been manipulated or staged. Over the weekend, Trump’s personal physician, naval officer Sean Conley, gave conflicting information about whether the president had been treated with oxygen. Conley said his reluctance to disclose the fact of Trump’s treatment with oxygen was an attempt not to “steer the course of illness in another direction.” Conley also misled — unintentionally, he said — about the timing of Trump’s diagnosis, a fact ultimately clarified by an October 4 report. It claims that the White House did not disclose the first positive coronavirus test when Trump called into Fox News’ “Hannity,” even though the test result was already known by then. 

Trump and his surrogates seem to be advancing, or at least abetting, misleading narratives in an apparent effort to paint a rosy picture of his condition. When Trump tweeted a video aiming to reassure the public about his health, observers quickly noticed an incongruity around the 1:04 mark, which some video editing experts say contains signs of digital manipulation, possibly to conceal a cough. Photos of Trump released by the White House, supposedly showing the president at work in two different locations, contain metadata that suggests they were taken ten minutes apart, prompting accusations they were staged and that Trump was in reality too sick to work. The cases demonstrate how fumbled public relations efforts can generate more rumors and conspiracy theories — some persuasive, others increasingly outrageous. Among the latter, Trump’s surprise motorcade drive-by Sunday led to questions about whether a body double had been used, including in one tweet shared over 1,800 times. That accusation wasn’t confined to Trump’s critics. In one QAnon-linked Facebook Group, some commenters wondered whether the president was really the person waving to his supporters, focusing on the supposedly unfamiliar way in which he was waving. 

Politicization of rape and murder case in India prompts misinformation and ‘plot’ claims

The politicization of the alleged rape and murder of a lower-caste Hindu woman in Hathras, Uttar Pradesh, has prompted hundreds of Twitter users to claim that expressions of outrage from the opposition Congress Party are a “plot” by the national opposition. There are more than 77,000 tweets using the hashtag #कांग्रेस_की_साजिश_हाथरस (“Hathras was the Congress’ plot”), many of which make unevidenced claims that the party’s leaders are trying to incite riots and violence, and are part of a conspiracy to “defame” the regional BJP government. A video of Uttar Pradesh’s chief minister claiming that “those that do not like development want to incite ethnic riots” was shared by the state government’s spokesperson using the hashtag. Some tweets with this hashtag falsely blame the rape on the Muslim community, despite the arrests of four Hindu men in connection with the crime. Others draw connections between the Congress party and “jihadi forces,” claiming that Congress’ protests could lead to “ethnic riots” financed by “money from Muslim countries and Islamic fundamentalist organizations” to deepen the social divide.” These sentiments echo the rising Islamophobia both in the mainstream media and online. 

Neck and neck (and neck and neck and neck…)

Be well,
The First Draft Team

First Draft Twitter
First Draft Facebook
First Draft Website
First Draft Youtube
First Draft Linkedin
Copyright © 2020 First Draft, All rights reserved.

Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.

Email Marketing Powered by Mailchimp
Think we missed something? Email