Twitter Files: The Removal of Trump, Employees Try to Justify Ban

DALLAS, TEXAS - AUGUST 06: Former U.S. President Donald Trump prepares to speak at the Conservative Political Action Conference CPAC held at the Hilton Anatole on August 06, 2022 in Dallas, Texas. CPAC began in 1974, and is a conference that brings together and hosts conservative organizations, activists, and world leaders in discussing current events and future political agendas. (Photo by Brandon Bell/Getty Images)Former U.S. President Donald Trump prepares to speak at the Conservative Political Action Conference CPAC held at the Hilton Anatole on August 06, 2022 in Dallas, Texas. (Photo by Brandon Bell/Getty Images)

OAN Newsroom Correspondent Sophia Flores
UPDATED 3:21 PM PT – Sunday, December 11, 2022

In this edition of “Twitter Files: The Removal of Donald Trump,” Michael Shellenberger focuses on what occurred in Twitter’s headquarters on January 7th. As pressure boiled, Twitter executives continued building the case for a permanent ban on the president. Within the span of a day, Twitter executives created justifications to ban Trump. In the midst of the events that occured on January 6th, the company introduced a policy change for Trump’s account. The new policy applied on the presidents account was distinctly different from policies other political leaders had. While doing this, executives expressed no concern for free speech or democracy implications of a ban.

For years, Twitter resisted calls to ban Trump. However, in the aftermath of January 6th, internal and external pressure on then Twitter CEO Jack Dorsey grew. After the events that took place at the Capitol, many notable figures, including Former First Lady Michelle Obama, voiced their opinions to ban the president from the platform.

But after the events of Jan 6, the internal and external pressure on Twitter CEO @jack grows.

Former First Lady @michelleobama , tech journalist @karaswisher , @ADL , high-tech VC @ChrisSacca , and many others, publicly call on Twitter to permanently ban Trump. pic.twitter.com/RzNj7WJReg

— Michael Shellenberger (@ShellenbergerMD) December 10, 2022

At the time, Dorsey was on vacation in French Polynesia. He phoned into meetings. However, he delegated much of the handling of the situation to senior executives Yoel Roth, the former Head of Trust & Safety, and Vijaya Gadde, the then Head of Legal, Policy & Trust.

It’s important to note that at the time, almost all of Twitter’s staffs political donations went to one political party. In 2018, 2019 and in 2022, 96%, 98% and 99% of their donations went to the Democrats.

On January 7th, Dorsey emailed employees reminding them that the social media company needed to remain consistent in its policies. Soon after his email, Roth privately messaged an employee reassuring them that “people who care about this, aren’t happy with where we are.”

On January 7, @Jack emails employees saying Twitter needs to remain consistent in its policies, including the right of users to return to Twitter after a temporary suspension

After, Roth reassures an employee that "people who care about this… aren't happy with where we are" pic.twitter.com/IfDpEVnOtR

— Michael Shellenberger (@ShellenbergerMD) December 10, 2022

In a different chat with his colleagues, Roth shared the news that Dorsey approved repeat offenders for civic integrity. The new approach would create a system where five violations, also known as strikes, would result in a permanent suspension. At this point in time, Trump had four strikes.

The excitement from his fellow employees in the chat made it clear that they had been pushing Dorsey for greater restrictions on the speech that Twitter allows around elections.

Around 11:30 am PT, Roth DMs his colleagues with news that he is excited to share.

“GUESS WHAT,” he writes. “Jack just approved repeat offender for civic integrity.”

The new approach would create a system where five violations ("strikes") would result in permanent suspension. pic.twitter.com/F1KYqd1Xea

— Michael Shellenberger (@ShellenbergerMD) December 11, 2022

One of Roth’s colleagues asked “does the incitement to violence aspect change that calculus?” Roth said it does not. The next day, Twitter announced the permanent ban on Trump’s account due to “further risk of incitement of violence.”

Roth's colleague's query about "incitement to violence" heavily foreshadows what will happen the following day.

On January 8, Twitter announces a permanent ban on Trump due to the "risk of further incitement of violence." pic.twitter.com/psLb5HDGQP

— Michael Shellenberger (@ShellenbergerMD) December 11, 2022

The only serious concern that Shellenberger found regarding implications for free speech and democracy of banning Trump, came from a junior level employee. It was discovered in a low-level Slack channel called “site-integrity-auto.”

"This might be an unpopular opinion but one off ad hoc decisions like this that don’t appear rooted in policy are imho a slippery slope… This now appears to be a fiat by an online platform CEO with a global presence that can gatekeep speech for the entire world…" pic.twitter.com/4pedmgY8pa

— Michael Shellenberger (@ShellenbergerMD) December 11, 2022

The term “one off” was often found in Slack conversations. Its frequent use revealed significant employee discretion over when and whether to apply warning labels on tweets and strikes on its users. Employees were aware of how their personal political opinions differed from Twitter’s Terms of Service.

Twitter employees recognize the difference between their own politics & Twitter's Terms of Service (TOS), but they also engage in complex interpretations of content in order to stamp out prohibited tweets, as a series of exchanges over the "#stopthesteal" hashtag reveal. pic.twitter.com/tfZesQNXx8

— Michael Shellenberger (@ShellenbergerMD) December 11, 2022

The hashtags “stopthesteal,” and “kraken” were added to a blacklist of terms to be deamplified. Roth’s colleague objected blacklisting “stopthesteal” because it risked “deamplifying counterspeech” that validates the election. The two quickly came up with a solution. They settled on deamplifying accounts that contained the phrase “stopthesteal” in its name or on its profile because those accounts “are not affiliated with counter speech.” The company ran into unexpected issues by blacklisting “kraken” because in addition to being a QAnon conspiracy theory based on the mythical Norwegian sea monster, it is also the name of a cryptocurrency exchange.

Another conundrum the company faced was whether or not to punish users who shared screenshots of Trump’s deleted January 6th tweets.

What if a user dislikes Trump *and* objects to Twitter's censorship? The tweet still gets deleted. But since the *intention* is not to deny the election result, no punishing strike is applied.

"if there are instances where the intent is unclear please feel free to raise" pic.twitter.com/8bdG6b38ej

— Michael Shellenberger (@ShellenbergerMD) December 11, 2022

The company also tried to decide what to do in cases where a user disliked Trump and objected to Twitter’s censorship. They decided that in that case, the tweet would get deleted. However, since the intention of the tweet was not to deny the election result, no strike would be applied.

Around noon on January 7th, a senior executive in advertising sales sent a private message to Roth asking what policies would need to be violated in order to permanently suspend Trump’s account. In response, Roth said that any policy violation would result in his suspension from the social media platform. The sales executive then went on to say “are we dropping the public interest (policy) now?” Roth responded by saying “In this specific case, we’re changing our public interest approach for his account.” This conversation is essential to understanding how Twitter justified banning Trump.

Twitter’s public interest exceptions allowed all elected and government officials to tweet anything that directly contributed to the understanding or discussion of a matter of public concern, even if their tweets violated Twitter’s terms. Trump was the only government official that was excluded from this rule.

The ad exec is referring to Twitter’s policy of “Public-interest exceptions," which allows the content of elected officials, even if it violates Twitter rules, “if it directly contributes to understanding or discussion of a matter of public concern” https://t.co/xTs14fD8V9 pic.twitter.com/ycbdlVmI7l

— Michael Shellenberger (@ShellenbergerMD) December 11, 2022

On the evening of January 7th, a Twitter engineer privately message Roth to let him know their worries about banning the president’s account. “I feel a lot of debates around exceptions stem from the fact that Trump’s account is not technically different from anybody else’ and yet treated differently due to his personal status without corresponding Twitter rules.” In response Roth hinted at how Twitter would justify deviating from its longstanding policy. “To put a different spin on it: policy is one part of the system of how Twitter works… we ran into the world changing faster than we were able to either adapt the product or the policy.”

Earlier that day, the employee wrote, "My concern is specifically surrounding the unarticulated logic of the decision by FB. That space fills with the idea (conspiracy theory?) that all… internet moguls… sit around like kings casually deciding what people can and cannot see." pic.twitter.com/KqwSdANBgo

— Michael Shellenberger (@ShellenbergerMD) December 11, 2022

That same employee spoke up one more time that night. “This might be an unpopular opinion but one off ad hoc decisions like this that don’t appear rooted in policy are imho a slippery slope and reflect an alternatively equally dictatorial problem. This now appears to be a fiat by an online platform CEO with a global presence that can gatekeep speech for the entire world.”

The next edition of the “Twitter Files: The Removal of Donald Trump” will show how Facebook’s suspension of Trump put Twitter in an awkward position. The pressure was on for Twitter to find a pretext to ban the 45th president. Bari Weiss, the founder and editor of The Free People, will release the unveil the final part of this series on Sunday, December 11th.

Twitter Files part 3: The suppression continues

SAN ANSELMO, CALIFORNIA - JANUARY 08: The suspended Twitter account of U.S. President Donald Trump appears on an iPhone screen on January 08, 2021 in San Anselmo, California. Citing the risk of further incitement of violence following an attempted insurrection on Wednesday, Twitter permanently suspended President Donald Trump’s account. (Photo Illustration by Justin Sullivan/Getty Images)The suspended Twitter account of U.S. President Donald Trump appears on an iPhone screen on January 08, 2021 in San Anselmo, California. (Photo Illustration by Justin Sullivan/Getty Images)

OAN Newsroom Correspondent Sophia Ariel
UPDATED 7:13 PM PT – Friday, December 9, 2022

The third installment of the Twitter Files was released on Friday. This time, the files are focusing on why 45th President Donald J. Trump was removed from the social media platform. This first part of ‘The Removal of Donald Trump’ thread covers October 2020 until January 6th.

Matt Taibbi announced that while talk of banning Trump from Twitter took place from January 6th until January 8th, the framework which led to his banning, was laid out months prior. Senior executives were using subjective moderation to choose what they believe did and didn’t violate the company’s policies.

A slack channel from the company titled “us2020_xfn_enforcement” showed the evolving thinking of top officials in late 2020 until early 2021. Through this channel, employees would discuss election related removals, especially tweets from high profile accounts.  During this time, executives were liaising with federal enforcement and intelligence agencies about moderation of election related content.

On posts the company deemed misinformation, Policy Director Nick Pickles was asked by his fellow employee if the company should state that Twitter detects “misinfo” through “ML, human review and partnerships with outside experts?” In response, Pickles said only the word “partnerships” should be used because he wasn’t sure if the FBI and DHS would be described as experts.

A post about the Hunter Biden Laptop situation is proof that Yoel Roth, the then Trust and Safety Chief, had weekly meeting with the FBI and the DHS.

The FBI would regularly send tweets that concerned them to the company.

While examining the election enforcement Slack, Taibbi was not able to find any references to moderation requests from the Trump campaign, the Trump White House, or Republicans. He was told that they exist, however they were absent from the channel.

One instance that started a long conversation in the Slack channel is when former Arizona Governor Mike Huckabee jokingly tweeted that he was filling out ballots for his deceased parents.

In response to his post, Twitter officials claimed that Huckabee was “literally admitting in a tweet to a crime.” The group declared Huckabee to be an “edge case.” However, they ultimately decided to leave him be. Roth suggested that even in a case such as Huckabee’s, where he was joking, the company needed to re-evaluate on what could be considered as misleading.

In order to combat the issue of tweets spreading misinformation, the company would put warning labels on tweets they deemed false.

Twitter was particularly strict on Trump. In one case, Trump quote tweeted an article about 50,000 Ohio voters who received wrong ballots. Immediately after the president posted his tweet, Twitter employees were ready to slap a “mail-in voting is safe” warning label on it. Shortly before adding the label, they realized that the tweet was factually accurate.

On October 26, 2020, Trump tweeted “Big problems and discrepancies with mail-in ballots all over the USA. Must have final total on November 3rd.” Even though senior executives did not find a particular violation in his tweet, they were fast to make sure it could not be “replied to, shared, or liked.”

There were many instances of pro-Biden tweets that warned Trump “may try to steal the election.” All of these anti-Trump tweets were approved by senior executives.

On December 10th, Twitter executives introduced the L3 deamplification tool. This announcement came while Trump was in the middle of firing off 25 tweets where he was commenting such things as “A coup is taking place in front of our eyes.” The deamplification was silently used to limit Trump’s outreach.

This tool is important to note, as it was deploying a vast range of visible and invisible tools to rein in Trump’s engagement long before January 6th.

Soon after the tool was put in place, bots were put on the 45th president’s personal Twitter account. In addition, the presidents account was bounced. When an account is bounced, it means that it is inaccessible for 12 hours.

Roth denied restricting the president’s account.

On January 6th, Trump tweeted to his supporters that they should “Go home with love & in peace.” His tweet did not go over well with the Twitter’s executives.

Vijaya Gadde, the former Head of Legal, Policy and Trust at Twitter, sent out her first company-wide email that day. In it, she announced that three of Trump’s tweets had been bounced. She also stated that his tweets signaled a determination to use legit “violations” as a guide for a possible permanent suspension.

The second installment of “The Removal of Donald Trump” will continue on Saturday, December 10th. Michael Shellenberger will detail the chaos inside Twitter on January 7th. The final installment of the series will be revealed by Bari Weiss on Sunday. That thread will reveal the secret internal communications from January 8th, the day Trump was removed from Twitter.