QArmyJapanFlynn (QAJF): The Collective Delusion Is Global

How Twitter allows fake accounts to create the illusion of widespread support

Network data (Following/Followers connections) for Japanese language QAnon accounts was collected by Social Forensics in April 2020 and spans thousands of accounts; a snapshot of ~80 accounts is presented above

In late July, Social Forensics penned a post detailing how Twitter’s announcement with regards to the QAnon collective delusion —namely, that they would be limiting the spread of QAnon-related content and accounts — was nothing more than lip service.

This post will focus specifically on Japanese language QAnon accounts, which Twitter’s part-time CEO, Jack Dorsey, continues to monetize, despite Twitter claiming to take a proactive approach towards mitigating against information operations/platform manipulation (and particularly those efforts connected to QAnon, having been linked to several acts of violence and dubbed a domestic terrorist threat by the FBI):

Screenshot captured September 17th, 2020; in addition to Twitter serving up a Rolex ad on the profile of @QAJF_bot, Twitter’s “You might like” algorithm encourages users to follow additional QAnon accounts (@zensen_t, @Sakamoto_Photo)

I bet Rolex will be thrilled to discover they are paying Twitter to have their brand appear adjacent QAnon (#WWG1WGA).

It is worth noting the link to QMap in the bio description of the @QAJF_bot account. QMap, prior to the site being taken offline last week and following an investigation by Logically, functioned as a hub for QAnon supporters to consume what are referred to as “Q drops” — which effectively are clues that have been left behind by the cult’s leader, Q.

Meet Eri (@okabaeri9111)

The @okabaeri9111 Twitter account — reflecting having nearly 80 thousand Followers — is the largest account among the Japanese language QAnon accounts:

Screenshot captured September 17th, 2020; similar to when viewing the profile of @QAJF_bot, Twitter’s “You might like” algorithm recommends following an additional QAnon account while viewing the profile of @okabaeri9111 (in this case, @yume_no_morix2)

According to the account’s bio description, Eri is the “sole [Japanese] translator” for QMap, in addition to being the “founder” of QAJFlynn (aka QArmyJapanFlynn). Moreover, the official QAJF site is listed in the account’s bio description, as are three Twitter accounts that, in our research, have emerged as being central to QAnon (@StormIsUponUs, @GenFlynn, and @intheMatrixxx). The @StormIsUponUs account, aka JoeM, was one of the largest QAnon supporting accounts prior to its suspension on April 9th, 2020. At the time of suspension, @StormIsUponUs reflected having 273 thousand Followers.

Given Eri(QMapJapan)’s Followers count is approaching 80 thousand, it would be easy for one to conclude that QAnon is wildly popular among Japanese speakers.

Upon closer inspection, however, it becomes clear that the vast majority of accounts following @okabaeri9111 are fake accounts.

Twitter’s refusal to enforce their own rules not only results in advertisers like Rolex wasting their advertising dollars on useless inventory (i.e. Rolex is a victim of blatant ad fraud where their brand appears adjacent QAnon), but also functions to create the illusion that QAnon has a massive following among Japanese speakers.

NOTE: Social Forensics has contacted Mounia Mechbal, Rolex’s VP of Marketing and Communications, to inform her that Twitter is engaging in ad fraud that presents Rolex’s brand adjacent QAnon (we will update this post should we receive a response):

Following & Followers Overlap

Of the 50,302 accounts that @okabaeri9111 was Following as of September 15th, 2020, 95% (47,756 accounts) were following the account back (i.e. Followers of @okabaeri9111).

Hence, it would be fair to categorize @okabaeri9111 as a follow-for-follow account.

More specifically, follow-for-follow accounts are generally Following a similar number of accounts (often a very large number, e.g. 50,302) as they have as Followers. Moreover, follow-for-follow accounts tend to be followed back by a large percentage of the accounts they are Following.

Most large follow-for-follow accounts have leveraged automation to create the illusion of support for their account/content (i.e. people are more likely to accept a tweet from an account with 100K Followers vs. a tweet from an account with 10 Followers as being a commonly held viewpoint, for example).

This tactic, which admittedly Twitter has more recently made adjustments to their policies to limit, was incredibly widespread for many years (and where it would be impossible for Twitter to be unaware of the downstream repercussions). Effectively it involves running a script (to simplify, this may be viewed as a piece of code) that automates the following process. For example, someone can compile (or acquire) a list of accounts that appear to be running a script which is set to automatically follow back any account that follows it. Next, they would simply set their script to follow said list of accounts, allowing them to very quickly have at their disposal an account that, on the surface at least, appears to have lots of support behind it (i.e. a large Followers count). In reality, the vast majority (but certainly not all) of accounts that run scripts which follow back any account that follows theirs are fake accounts. There is zero incentive for Twitter to remove these fake accounts, as many of them tweet, in turn creating advertising inventory, and thus revenue, for Jack Dorsey’s company.

The fact that 95% of the accounts being followed by Eri happen to be accounts that are following Eri back is something that simply does not occur organically.

It is worth noting that Twitter Rules prohibit “aggressive following”:

When Twitter suspends the account of @okabaeri9111 (which is inevitable), it is paramount that they also suspend the fake accounts inflating Eri’s Followers count. Else, chances are another Eri type of account will be created by the same operators, who will in turn backfill the account’s Followers with many of the same fake accounts, once again succeeding in creating the illusion of support.

Twitter tends to only remove a handful of core accounts while claiming to proactively mitigate information operations/platform manipulation, rather than removing the full ancillary networks that are being utilized by nefarious actors.

Followers Counts & Following/Followers Ratios

Below is a table that summarizes the Followers counts and the Following/Followers ratios of the 76,736 accounts that were following @okabaeri9111 (i.e. Eri’s Followers) as of September 15th, 2020:

Nearly one-third of @okabaeri9111’s Followers (24,732 accounts) are accounts that have fewer than 100 Followers.

There are 12,537 accounts, for example, that follow @okabaeri9111 which have between 5,000 and 39,999 Followers. Meanwhile, there are 7,164 accounts that follow @okabaeri9111 which have between 1,000 and 1,999 Followers.

The 12,537 accounts (with Followers between 5,000 and 39,999), on average, reflect having Following counts that are the same as Followers counts (ratio = Following/Followers; avg ratio = .98).

The 7,164 accounts (with Followers between 1,000 and 1,999), on average, reflect having Following counts that are 149% of Followers counts (avg ratio = 1.49).

Effectively, this table illustrates that the vast majority of Eri’s Followers are follow-for-follow accounts that are in violation of Twitter Rules.

Verified Accounts

Of the 76,736 accounts that follow @okabaeri9111 just 55 verified accounts emerge. Included among them are former National Security Advisor Michael Flynn, National Tea Party movement co-founder, Michael Johns, and 18 current or former QAnon supporting 2020 congressional candidates:

Yellow rows = current or former QAnon supporting 2020 congressional candidates

The earliest current or former QAnon supporting congressional candidate to follow Eri was patriot Duane Hennen of Ohio (who was the 1,463rd account to follow okabaeri9111):

As may be seen above, Mr. Hennen has cultivated more than 30 thousand Followers via implementing the follow-for-follow tactic that was described earlier in this post (and which violates Twitter Rules)

National Tea Party movement co-founder, Michael Johns, who follows a whopping 153 thousand accounts, was the 3,800th account to follow Eri:

Johns, who Social Forensics has written about previously, plays a central role in the Tea Party’s well-oiled Twitter propaganda machine:

When confronted via Twitter DM re: his thoughts on social media manipulation, Johns stated he was against it.

Eri’s Patreon

The link field of @okabaeri9111’s Twitter profile points to Eri’s Patreon page (userid=19036219):

Eri offers two membership levels ($1 per month or $5 per month) and currently has 233 patrons paying for “creating Translate Qmap.pub for Japanese.”

As a reminder, the QMap site was taken offline last week following an investigation by Logically.

OK, So What’s Your Point

Twitter’s part-time CEO, Jack Dorsey, continues to monetize ad inventory created by QAnon accounts that violate Twitter Rules, and despite Twitter recently (July 21st, 2020) asserting that they “will stop recommending accounts and content related to QAnon, including material in email and follow recommendations, and [that they] will take steps to limit circulation of content in features like trends and search.”

Additionally, as was illustrated in this post, Twitter’s “You might like” algorithm regularly recommends additional QAnon accounts to follow when viewing the profiles of accounts that Twitter has committed to removing from their platform. Twitter’s account recommendation algorithm is incredibly problematic in that in encourages people to go further down the QAnon rabbit hole.

Perhaps Dorsey is himself a QAnon supporter, and hence refuses to enforce the Twitter Rules? 🧐🤨🤔

Part-time Twitter CEO, Jack Dorsey, and Twitter Legal, Policy and Trust & Safety Lead, Vijaya Gadde, meet with President Trump (April 2019)

Social Forensics maps and monitors social connections and activity.

We create purposefully designed tools to manage social data analytics needs across various industries. Our focus is audience segmentation and identifying coordinated inauthentic behavior (CIB) across social media platforms.

Geoff Golberg is an NYC-based researcher (and entrepreneur) who is fascinated by graph visualization/network analysis — more specifically, when applied to social networks and blockchain activity. His experience spans structured finance, ad tech, and digital marketing/customer acquisition, both at startups and public companies.

Geoff is the Founder/CEO/Janitor of Social Forensics.

CEO & Founder, Social Forensics | Previously: Co-Founder, Elementus | Featured in BBC, CNN, BuzzFeed, and Quartz, among others | SocialForensics.com

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store