Harris Campaign Caught Manipulating X To Censor Criticism

In part one of this investigation into how Kamala Harris’ presidential campaign is deceptively manipulating online platforms, it was revealed that the campaign is operating a Discord server that directs hordes of volunteers to use their social media accounts to deceptively push election propaganda.

The goal is to artificially manufacture consensus by making pro-Kamala Harris messages on social media appear more popular than they are, and it is often done in violation of the Terms of Service of the social media platforms. In more extreme cases, they are encouraging people to skirt election laws and using these “astroturfing” campaigns to spread disinformation they think will help win the upcoming election.

Yesterday’s report documented how the Harris-Walz campaign has seen great success in manipulating Reddit’s algorithm, but that isn’t the only social media site they’re manipulating. The campaign has also been targeting Elon Musk’s X, perhaps the most influential site for political news. One particular goal, according to a user of the Harris-Walz campaign Discord server, is to get campaign volunteers to swarm the site and “block [community notes] we don’t like.”

Prior to Musk’s purchase of Twitter, the site’s management was known for capriciously removing information and regularly banning users in a way that employed a double standard that heavily disfavored conservative opinion. Musk, a major free speech advocate, sought to institute a more neutral way to deal with misleading tweets, and the “Community Notes” system was born.

Select users who signed up for the program could propose notes to be added to tweets showing that the information was wrong, misleading, or required important context. Other users can then read the proposed notes and vote on whether they are accurate or needed, and if the proposed notes get enough favorable votes, they get appended to the post permanently.

However, throughout this campaign Harris-Walz official accounts have been remarkable conduits for disinformation and have regularly provided dishonest presentations of the Trump-Vance campaign. The Harris campaign’s X accounts have been so bad that even CNN fact checker Daniel Dale, who has been very harsh on Trump for several years, wrote an entire column highlighting the errors and dishonesty.

Despite this, virtually none of the false and misleading tweets from Kamala Harris’ campaign have Community Notes appended to them. One likely explanation for this discrepancy is that the Harris campaign is directing volunteers on its Discord server to vote down Community Notes even when those notes accurately say the campaign is being deceptive.

In the example below, after a @KamalaHQ tweet claimed Trump referred to Americans who don’t support him as “dangerous people,” Timothy Durigan, an employee of the Democratic National Committee, urges campaign volunteers to vote down a Community Note that accurately pointed out the Kamala campaign was taking Trump’s remark out of context. Trump was actually speaking about those in the American government responsible for leaking information related to Israel’s war plans:

In the “twitter-community-notes” channel found on the Harris-Walz Discord, paid Democrat staffers are also writing dubious Community Notes on X to undermine GOP and Trump messaging. They then encourage volunteers to rate them positively.

Unsurprisingly, these notes are often filled with half-truths, misleading information, or lies, such as explaining how Joe Biden’s son’s brain tumor must have been due to his service in Iraq.

The Harris-Walz server even has a “Twitter (X) Community Notes Training” module, which describes how users can quickly bump up their Community Notes “Rating Impact,” which allows them to write their own Community Notes with a high enough status. They also describe “problems” with Community Notes, such as Joe Biden being tagged in “inaccurate Community Notes,” and how users can rate them negatively.

Since X’s Community Notes upvoting system is designed to mitigate political bias far better than Reddit’s, the Harris-Walz campaign’s attempt to manipulate Community Notes on X hasn’t been as successful.

One Harris-Walz Discord user lengthily expressed his frustrations at the bias mitigation system used by Community Notes (CNs), which was implemented to prevent user manipulation:

While their attempts to abuse Community Notes on X were largely ineffective, it is still a gross violation of X’s Terms of Service, which prohibits artificially amplifying information.

Other Astroturfing Operations

The Harris-Walz Discord server provides access to an app called “Reach” which gives its users access to a database of “entertaining” Harris-Walz campaign content.

X has also introduced “Radar,” a feature that lets users see post volumes on specific topics. Using messaging guidance from three Reach posts, I found evidence of hundreds of astroturfing profiles for the Harris campaign on X. The same search was done on Google with one of the images, suggesting broader reach. While the Google data is larger, it is also likely incomplete while the X data is complete. The scale of this operation continues to grow.

Users can connect their own personal social media accounts to the Reach app, so they can easily repost memes, videos, and other content that promotes the ideals of the Harris-Walz campaign.

While topics relating to current events routinely trend on Reddit and X, other platforms, such as TikTok and Instagram, gear themselves towards entertaining content. To target these platforms, the campaign is also using Reach. Unfortunately for them, the content they provide pretty much resembles your liberal aunt’s Facebook timeline.

To people who have spent years online learning the nuances of meme culture, this is fundamentally repulsive. It embodies the inauthentic nature of the modern internet, which has seemingly replaced the “old net” in the last two decades.

Their reaction to the now infamous Man Enough to Vote for Harris ad speaks for itself:

Not only do they love flamboyant actors masquerading as American “men,” they also enjoyed actively promoting it during a football game, boosting their post with unrelated hashtags that happened to be trending at the time. This is yet another shady tactic used to bump up social media posts inauthentically, shamelessly employed by official Harris-Walz volunteers.

But at this late stage in the campaign, it would appear spreading inauthentic content and deceptive messages is all the Harris campaign can do.  

The author runs the popular Twitter account @reddit_lies.