Twitter Files Part 2: Twitter and its shadow bans
The company has already been doing what Musk said it will
➡️ The Shortcut Skinny: Twitter Files Part 2
👤 New 30-tweet series show Twitter’s internal “shadow ban” tools
🚫 Alleges biased enforcement of Twitter’s harmful content rules
🤥 Former Twitter top brass said it doesn’t shadow ban, even though it does
📜 Elon Musk advocated “freedom of speech, not freedom reach” recently
⏪ Musk now says users will be notified of bans, told why, told how to reverse
Twitter Files Part 2
Our Twitter Files explainer is expanding, now that new revelations – dubbed Twitter Files Part 2 – purportedly show Twitter’s internal “shadow banning” tools, which the report alleges were disproportionately used on notable right-wing personalities.
The latest batch of “Twitter Files” come from former New York Times columnist Bari Weiss who, in a series of 30 tweets, gave us our first look at internal screenshots showing various ways prominent users of the platform have had their reach limited.
Twitter Files Part 2 follows a Friday, December 2 flurry of tweets from journalist Matt Taibbi, who was given the documents to cover and analyze, which largely revealed government requests around Hunter Biden’s laptop and the dissemination of information surrounding that. Our Twitter Files explained article has a timeline analysis of that story if you want to catch up.
So who was shadowbanned in Twitter Files Part 2?
Dr. Jay Bhattacharya, a professor of medicine at Stanford University, was marked as being on a “Trends Blacklist” and as having a “Recent Abuse Strike.” Fox News host Dan Bongino was placed on a “Search Blacklist,” making his account difficult to find within Twitter’s search tool. Conservative activist and Turning Point USA founder Charlie Kirk was marked “Do Not Amplify,” presumably meaning his tweets wouldn’t be included in the algorithms that normally promote certain users.
The Twitter account Libs of TikTok had several flags saying things like “Trends Blacklist,” “Recent Abuse Strike,” and “Strike Count: 2,” and even had a red tag at the top that read, in all caps, “Do not take action on user without consulting with SIP-PES@.”
The images are presented largely without context, but each account had a history of running afoul of Twitter and other social media companies’ rules surrounding misinformation, at least with the old upper management of Twitter.
Dr. Bhattacharya participated as an expert in a YouTube COVID-19 roundtable discussion with Florida Governor Ron DeSantis, Dr. Sunetra Gupta of Oxford University and Dr. Martin Kulldorff of Harvard medical school that was removed for containing content that contradicted health authority consensus on the efficacy of masking to prevent COVID-19 spread.
Mr. Bongino also ran afoul of the Google-owned video site, and was permanently banned from YouTube in January 2022 and stripped of the ability to earn Google ad money. Charlie Kirk was suspended from Twitter in March 2022 after misgendering Rachel Levine, the US assistant secretary of health. Libs of TikTok has routinely been accused on the platform of amplifying hateful anti-trans rhetoric, and has seen suspensions as a result, though the account maintains it’s just reposting TokTok videos in their own words
Weiss’s Twitter Files Part 2 thread further showed an internal memo that acknowledged Libs of TikTok hadn’t directly violated Twitter’s hateful conduct policy and therefore could not be permanently banned, nor would any number of temporary suspensions lead to such a ban. When Chaya Raichik, who owns the account, complained to Twitter after a photo of her home and address was posted in a tweet – essentially doxxing – Twitter responded that the tweet wasn’t in violation of its rules, and no action would be taken.
I attempted to locate the tweet in question, and found several from November 20, 2022 that may have been it, but there is some question about whether the address posted was Raichik’s home address or the business address, possibly of the real estate firm for which she works.
Is Twitter Files Part 2 a bombshell?
It depends on who you ask. The fact that Twitter was shadowbanning isn’t surprising. It’s been alleged by people – mostly conservatives – for years, without proof, that it was being used selectively based on political alignment. Apart from the implication of bias though, it seems to generally fall within the company’s policies.
Prior to this week’s Twitter Files Part 2, we have lacked any visual of Twitter’s shadow-banning tool, the existence of which flies in the face of what has been said by Twitter executives, who seemed to have the final say in the execution of the tool, at least for certain high-profile accounts.
Kayvon Beykpour, former head of product at Twitter, tweeted in 2018: “We don’t “shadow ban.” The tweet linked to a Twitter blog post clarifying the company’s definition of shadow banning as being “deliberately making someone’s content undiscoverable to everyone except the person who posted it, unbeknownst to the original poster.”
That Twitter didn’t shadow ban is misguided at best and intentionally misleading at worst.
With a strict interpretation of that definition in hand, sure, technically none of the above constitutes shadow banning, since the content was still discoverable in one way or another, but it’s a convenient distinction that isn’t universally accepted as the full meaning of the term. The statement, then, that Twitter didn’t shadow ban is misguided at best and intentionally misleading at worst.
Vijaya Gadde, former head of legal, policy and trust, quoted that tweet, expanding: “Twitter exists to serve the public conversation, enabling important discussions around the world to occur. Favoring one specific ideology or belief goes against everything we stand for.”
That stands in stark contrast to what we’re seeing from Twitter Files Part 2. That said, it does seem to mirror Elon Musk’s own tweet from November 18 of this year stating that “negative/hate tweets will be max deboosted & demonetized,” making them only findable if you do so intentionally, which he says “is no different from the rest of the internet.”
The difference, which only came to light today, is that Musk plans to show users their “true account status, so you know clearly if you’ve been shadowbanned” and allow for a way to appeal this restriction.
Twitter Files Part 3?
Weiss promises further exposé on her website, so hopefully there will be more relevant context there. As of today, the information presented raises plenty of questions while answering few, if any. For instance: how many times were these consequences carried out on folks of other political persuasions than what’s presented to us? The implication seems to be that it never was, but with the only arbiters of the data being a few hand-selected journalists, that’s harder to see.
Lastly, there was some question initially that Weiss’ tweets meant she and Taibbi have direct access to Twitter’s systems. The current head of trust and safety at Twitter, Ella Irwin, says that’s not the case – she provided the images.
