Twitter Files Redux. Parts 6 through 10
Discover more from Rational Ground by Justin Hart
Twitter Files Redux. Parts 6 through 10
In case you missed parts 1 through 6 —> see here.
As a note of context: I have 8 kids and a full-time job but I also have a wife who is meticulous and an excellent writer - she’s also honest to a fault and encouraged me to admit that I did use some article intelligence to help me summarize these massive tweet threads! It’s the way of the future folks!
Twitter Files - Part 6
At the very least the Twitter files have shed new light on the extent of the US government's monitoring of social media platforms, and in particular, Twitter. The documents in this next batch (Part 6) presented by Matt Taibbi, which span from January 2020 to November 2022, reveal that Twitter had a constant and pervasive relationship with the FBI, with over 150 emails exchanged between former Twitter Trust and Safety Chief Yoel Roth and the FBI during this time period. The emails show the FBI requesting action on election misinformation, even including joke tweets from low-follower accounts, as part of the FBI's social media-focused task force, known as FTIF. This task force, created in the wake of the 2016 election, included 80 agents and corresponded with Twitter to identify alleged foreign influence and election tampering.
The Department of Homeland Security (DHS) also partnered with security contractors and think tanks to pressure Twitter to moderate content. It is not clear from the documents whether agencies such as the FBI and DHS carried out their own flagging work or outsourced it. However, the sheer volume of reports from the government that were sent to Twitter through multiple entry points, pre-flagged for moderation, is notable.
In one email from November 2022, the FBI requested action on four accounts, including @fromMA, whose tweets were mainly jokes. Twitter personnel went on to look for reasons to suspend the accounts, with only two – @ClaireFosterPHD and @fromMA – avoiding suspension. In another internal email from November 5, 2022, the FBI's National Election Command Post compiled a list of accounts that "may warrant additional action" and sent it to the San Francisco field office. The field office then forwarded the list to its "Twitter folks", who responded with a list of actions taken. Many of the accounts on the list were suspended, with Twitter responding with a generic "Thanks, Twitter" letter. Just insane, really.
The Twitter Files also show that the FBI requested access to the company's internal moderation tools and sought to influence Twitter's content moderation policies. In a letter to former Deputy General Counsel Jim Baker (the spook on the payroll) on September 16, 2022, the FBI requested access to Twitter's "shadowbanning" tools, which allow the platform to quietly reduce the visibility of certain accounts. The FBI argued that access to these tools would allow them to better track and combat online child sexual exploitation. Twitter declined the request - thankfully.
The documents also reveal that Twitter was asked to consider implementing a "timeout" feature, which would allow users to take a break from the platform without deleting their account. The feature was requested by the DHS, who argued that it would be useful in combating terrorism and extremism. However, the feature was not implemented - thankfully.
The Twitter Files provide insight into the extent of government monitoring of social media platforms and the pressure that these platforms face to moderate content. They also raise questions about the role of these platforms in combating misinformation and the potential for government overreach in regulating online speech. While Twitter has faced criticism for its handling of misinformation and abuse on its platform, the Twitter Files show that the company has also pushed back against requests from the government that it views as inappropriate or overreaching.
Twitter Files - Part 7
In December 2019, the owner of a computer store in Delaware named John Paul Mac Isaac contacted the FBI about a laptop that had been left with him by Hunter Biden. The FBI issued a subpoena for the laptop and took possession of it. However, by August 2020, Mac Isaac still had not heard back from the FBI and discovered evidence of criminal activity on the laptop. He reached out to Rudy Giuliani, who was under FBI surveillance at the time, and in October 2020, Giuliani gave the information to The New York Post.
On October 13, 2020, just hours before the Post was set to publish an article based on the information from the laptop, Hunter Biden’s lawyer contacted Mac Isaac. The next day, the Post published an article revealing Hunter Biden’s business dealings, which were found to be accurate. However, within hours of the article’s publication, Twitter and other social media platforms censored the article, undermining its credibility in the minds of many Americans.
It has since come to light that there was an organized effort by the intelligence community to influence Twitter and other platforms in order to discredit the information about Hunter Biden before and after it was published. This included prompting Twitter’s Head of Site Integrity, Yoel Roth, to dismiss reports about the laptop as a Russian “hack and leak” operation. The intelligence community also influenced Facebook in a similar manner, according to CEO Mark Zuckerberg.
It is unclear exactly what the motivations were behind this effort to discredit the information about Hunter Biden. However, it is clear that there was an organized effort to interfere with the spread of accurate information and to mislead the public. This raises serious concerns about the role of the intelligence community in shaping public discourse and the extent to which social media companies are willing to censor information at the behest of government agencies.
Twitter Files - Part 8
Twitter has been accused of directly assisting the US military's online psychological influence campaigns, despite the social media giant's claims that it makes efforts to detect and thwart government-backed platform manipulation. In 2017, a US Central Command (CENTCOM) official sent Twitter a list of 52 Arab language accounts that were being used to amplify certain messages, and asked for priority service, verification and "whitelist" abilities for some of them. The same day, Twitter officials used a tool to grant a special "whitelist" tag to the accounts, which essentially provided them with verification status without the blue checkmark, meaning they were exempt from spam and abuse flags and more likely to trend on hashtags.
The CENTCOM accounts tweeted frequently about US military priorities in the Middle East, including anti-Iran messages, promotion of the Saudi Arabia-US-backed war in Yemen, and "accurate" US drone strikes that claimed to only hit terrorists. However, CENTCOM then changed strategy and deleted disclosures of ties to the Twitter accounts, with the bios of the accounts being changed to appear more organic. Many of these secretive US military propaganda accounts continued tweeting through this year, with some not being suspended until May 2022 or later.
In August 2022, a report from the Stanford Internet Observatory exposed a US military covert propaganda network on Facebook, Telegram, Twitter and other apps using fake news portals and deep fake images and memes against US foreign adversaries. The propaganda network was found to be pushing narratives against Russia, China, and other countries, and accused Iran of "threatening Iraq's water supply" and "plotting to kill US soldiers".
Twitter has faced criticism for not taking action against these accounts, with some Twitter officials reportedly feeling deceived by the covert shift. Emails from 2020 show that high-level Twitter executives were aware of the Pentagon's network of fake accounts and covert propaganda, but did not suspend them. In one email, Twitter lawyer Jim Baker discussed an upcoming meeting with the Department of Defense (DoD), stating that the Pentagon had used "poor tradecraft" in setting up its network and was seeking strategies for not exposing the accounts "linked to each other or to DoD or the USG".
Twitter has pledged to rapidly identify and shut down all state-backed covert information operations and deceptive propaganda, but it appears that the social media giant has not fully followed through on this promise. More on this to come I’m sure.
Twitter Files - Part 9
Over the past few weeks, reports have been published detailing the close coordination between the FBI and Twitter in moderating social media content. These reports, known as the "Twitter Files," reveal the FBI's role in facilitating a vast program of social media surveillance and censorship that encompasses agencies across the federal government, including the State Department, the Pentagon, and the CIA. The operation is larger than the 80 members of the Foreign Influence Task Force (FITF), which also facilitates requests from a wide range of smaller actors, such as local law enforcement, media, and state governments.
According to the Twitter Files, the FBI had so much contact with so many different agencies that Twitter executives struggled to keep track. This resulted in thousands of official reports being sent to Twitter from across the government through the FITF and the FBI's San Francisco field office. In one email, an FBI agent asked if he could invite an "OGA," or "Other Government Agency," to an upcoming conference. "OGA" can be a euphemism for the CIA, according to former intelligence officials and contractors.
The Twitter Files also show that the government was in constant contact not only with Twitter, but with virtually every major tech firm, including Facebook, Microsoft, Verizon, Reddit, and even Pinterest. Industry players also held regular meetings without government involvement. One common forum for these meetings was the FITF, which was attended by executives, FBI personnel, and nearly always one or two attendees marked as "OGA."
Despite its official mandate of "Foreign Influence," the FITF and the SF FBI office often received requests for domestic moderation from state governments and even local police. These requests often arrived through Teleporter, a one-way platform where communications are timed to disappear. As the 2020 election approached, the FITF and the FBI inundated Twitter with requests, sending lists of hundreds of problem accounts.
The FBI also appears to have tailored its searches to Twitter's policies and provided feedback to the platform on how to improve its moderation processes. Additionally, the Twitter Files reveal that the FBI and Twitter worked together to develop new tools and features to help identify and remove problematic content.
One major concern raised by the Twitter Files is the potential for government overreach and abuse of power in social media moderation. The FITF and the SF FBI office often received requests for the removal of content that was protected by the First Amendment, raising questions about the potential suppression of free speech. Additionally, the close relationship between the government and tech companies could create conflicts of interest, as the companies may be more inclined to comply with government requests in order to maintain favorable relationships.
Overall, the Twitter Files highlight the significant role that the government plays in social media moderation and the potential for abuse of power in this realm. Censorship is reall y’all.
Twitter Files - Part 10
We were thrilled to learn that David Zweig got the call to do the next set of Twitter Files - on Covid! Our coverage on this one goes deeper here.
Short version: both the Trump and Biden administrations had pressured Twitter and other social media platforms to elevate certain content and suppress other content about Covid-19. The article states that at the onset of the pandemic, the Trump administration was particularly concerned about panic buying and sought help from tech companies to combat misinformation about grocery store shortages. The Biden administration, on the other hand, focused on "anti-vaxxer accounts," specifically targeting former New York Times reporter, Alex Berenson, who was suspended from Twitter hours after President Biden commented that social media companies were "killing people" for allowing vaccine misinformation. Berenson later sued and settled with Twitter, with the company releasing internal communications that showed direct White House pressure to take action against Berenson.
The article notes that Twitter's moderation of content related to Covid-19 was problematic in three ways. Firstly, much of the moderation was conducted by bots that were trained on machine learning and AI, but were not sophisticated enough to handle nuanced content. Secondly, contractors in places like the Philippines were given decision trees to assist with moderation, but may not have had the expertise to accurately adjudicate tweets on complex topics. Finally, higher level employees at Twitter subjectively decided on escalated cases and suspensions, which may have resulted in bias.
Zweig states that Twitter censored information that was true but inconvenient to U.S. government policy, discredited doctors and other experts who disagreed with official positions, and suppressed ordinary users, including those sharing data from the Centers for Disease Control and Prevention (CDC). The article suggests that this suppression of certain viewpoints limited the public debate on Covid-19 and may have had serious consequences for public health.
2 more and hopefully we’ll have some new stuff to drop for you!
Twitter Files Redux. Parts 6 through 10
thank you