Written by Nick Heer.

Twitter’s New PR Team Finds More Normal Behaviour to Spin Conspiracies Around

Elon Musk may have eliminated Twitter’s communications department. But it appears he found a new PR team member in Bari Weiss, to whom the so-called “Twitter Files” were leaked, and who will massage Twitter 2.0’s messaging for free. Twitter apparently added Weiss to the company’s Slack channels and gave her a company laptop. You know, just a typical reporter–source relationship.

Weiss, today, tweeted a thread all about Twitter’s apparently shocking policy of minimizing the visibility of some high-profile and often controversial accounts:

What many people call “shadow banning,” Twitter executives and employees call “Visibility Filtering” or “VF.” Multiple high-level sources confirmed its meaning.


“VF” refers to Twitter’s control over user visibility. It used VF to block searches of individual users; to limit the scope of a particular tweet’s discoverability; to block select users’ posts from ever appearing on the “trending” page; and from inclusion in hashtag searches.

Weiss showed screenshots of a few users who have been filtered for various reasons, including Jay Bhattacharya — who used his doctorate status to argue against virtually all COVID containment measures — and Dan Bongino and Charlie Kirk, who promoted conspiracy theories about the 2020 U.S. election. To be clear, all of these users still have Twitter accounts and they are still able to tweet. Their millions of combined followers still see their messages. But their accounts may not autocomplete in search results or their tweets may not be promoted in trending topics.

All of these users happen to promote views typical of an American conservative and even far-right ideology, but it is impossible to know whether this is an accurate representation of the accounts which are flagged, despite what Musk says. We do not even know which specific tweets caused Twitter to flag these accounts; maybe they deserved it. Weiss is not a reliable narrator, and not just because she appears to have attained a volunteer consultancy role in the Musk era of Twitter. Weiss spent over an hour writing thirty tweets without once mentioning that Musk’s own policy position favours reducing “freedom of reach” for “negative” tweets. “You won’t find the tweet unless you specifically seek it”, said the guy who gave Weiss and Taibbi a bunch of internal documents from the site he runs. This appears to be a similar policy to the one Weiss spent an hour exposing as some kind of massive controversy when it was done under previous management.

Weiss’ big reveal was what appeared to be a contradiction between Twitter’s past stance and its actions:

Twitter denied that it does such things. In 2018, Twitter’s Vijaya Gadde (then Head of Legal Policy and Trust) and Kayvon Beykpour (Head of Product) said: “We do not shadow ban.” They added: “And we certainly don’t shadow ban based on political viewpoints or ideology.”

Again, Weiss has not provided evidence to indicate a political or ideological motive. You are supposed to draw the conclusion she has suggested based on her specific framing.

Here, Weiss trims the full context of Gadde’s statements by not linking to the post in question:

People are asking us if we shadow ban. We do not. But let’s start with, “what is shadow banning?”

The best definition we found is this: deliberately making someone’s content undiscoverable to everyone except the person who posted it, unbeknownst to the original poster.

We do not shadow ban. You are always able to see the tweets from accounts you follow (although you may have to do more work to find them, like go directly to their profile). And we certainly don’t shadow ban based on political viewpoints or ideology.

Not only do Gadde and Beykpour deny that Twitter shadow bans, they define their understanding of shadow banning, which is very different from the behaviour Weiss documents. None of the types of “Visibility Filtering” shown match the definition in the blog post above. (Update: And two former Twitter employees say “Visibility Filtering” is misdefined in Weiss’ thread.) Weiss is counting on people to not go looking for the full context because it would undermine her argument.

By the way, Gadde and Beykpour go on to describe how this process works:

We do rank tweets and search results. We do this because Twitter is most useful when it’s immediately relevant. These ranking models take many signals into consideration to best organize tweets for timely relevance. We must also address bad-faith actors who intend to manipulate or detract from healthy conversation.

They explicitly state Twitter down-ranks tweets in places like search “from bad-faith actors who intend to manipulate or divide the conversation”. You might disagree with specific decisions Twitter has made — so do I — but I find it hard to be even remotely upset by this standard. Just because a tweet is popular and comes from a user with millions of followers, it does not mean most users should be subjected to it. That is particularly true when it comes to health and democracy.

The last series of tweets in Weiss’ thread concern the hate-filled Libs of TikTok account, run by Chaya Raichik. Raichik’s posts frequently misrepresent drag events and target trans youth. Tweets from the account repeatedly traffic lies about LGBTQ people and topics, including tying it to pedophilia. This account has been suspended several times for violating Twitter’s policies against deliberately misgendering people. Raichik’s posts are intentionally and repeatedly hostile.

Weiss claims an internal memo confirms “[Libs of TikTok] has not directly engaged in behavior violative of the Hateful Conduct policy” but, again, this statement has been taken out of context. I would like to believe this is a simple error by Weiss but, as she has repeatedly done so in this thread to make her arguments — to say nothing about her entire body of work — it is hard to believe that is the case. The memo, of which Weiss posted a screenshot, says of the Libs of TikTok account:

Since its most recent timeout, while LTT has not directly engaged in behavior violative of the Hateful Conduct policy, the user has continued targeting individuals/allies/supporters of the LGBTQIA+ community for alleged misconduct.

The phrase “since its most recent timeout” is relevant context omitted by Weiss, as is the site policy team’s assertion that Raichik continues to harass and discriminate. This team concludes Raichik’s account deserves another time-out period based on her repeatedly offensive behaviour which, the memo implies, should indicate that her tweets are unacceptable and could eventually lead to a full ban. I do not see anything controversial here. If anything, Twitter is being lenient: an administrative view explicitly marks Libs of TikTok as a high profile account, and requires someone to consult an elevated level of Twitter policy maker before taking action. Weiss portrays this elevated policy team as a “secret group” but, well, it is disclosed right there in the admin view. Not a very well-kept secret, is it?

Weiss sets up a comparison between the apparently awful treatment of Raichik’s account and a lack of enforcement against a tweet which apparently included a picture of her house and its address. I wrote “apparently” there because I cannot find the tweet despite Weiss saying it remains live. It was allegedly posted just a couple of weeks ago, meaning the call to permit the tweet was made under Musk’s ownership and responsibility. Maybe all of those staffing cuts have made it more difficult to keep up with reported tweets. In any case, I do not think it was a fair call: if that tweet was as described, it should have been removed, but Libs of TikTok should have been banned long ago for bullying and inciting harassment, including posting private information.

Weiss may have rambled for an hour in a Twitter thread and her team may have been given ridiculous access (Update: that access has been disputed by Twitter’s head of trust and safety which is about what I expected for the reliability of this thread), but this is yet another apparently blockbuster exposé which has turned up little of note. Am I supposed to be surprised that accounts which traffic in bad-faith narratives are de-emphasized? I thought the whole thing was “freedom of speech doesn’t mean freedom of reach”?

Mike Masnick of Techdirt published an excellent and thorough article about the many misconceptions of the first “Twitter Files” thread. It is worth reading if you care about this sort of thing — hey, you are at the bottom of my post about this, so maybe you do — but I think his conclusion is reusable for this iteration of this saga:

I fear that this story is going to live on for years and years and years. And the narrative full of nonsense is already taking shape. However, I like to work off of actual facts and evidence, rather than fever dreams and misinterpretations. And I hope that you’ll read this and start doing the same.

Weiss’ thread today involved even more frequent excursions from truth and complete context despite being part of Twitter’s internal systems. I never want to hear a critical word about “access journalism” from anyone who has promoted this thread. I wish this did not feel like a big story. But it will be treated like one because it has a juicy combination of internal documentation, moderation policy, and the whines of an apparently oppressed real estate agent turned anti-transgender propagandist. Truly, the real victim in all of this.

If you are angry about the silent de-emphasis of some accounts, I hear you; I think more transparency could be useful. But I think most people also know when they are doing things which get right up to the line and test platform moderators’ patience. Some people are just assholes. And some people believe it is their duty to run interference for them by tweeting in spooky undertones about normal decisions to figure out how much of an asshole someone is being.