MSNBC’s Mehdi Hasan Gets Basic Facts Wrong on DHS Content Moderation Partnership

Exclusive Twitter Files emails and publicly available documents show Mehdi Hasan's accusation is false.

MSNBC pundit Mehdi Hasan, in a series of tweets over the last five days, has repeatedly claimed that Matt Taibbi “deliberately & under oath misrepresented” the facts when he testified to Congress last month. 

Hasan linked to a video of Taibbi's testimony, in which the Racket News journalist noted that "Twitter executives did not distinguish between [the Department of Homeland Security] or [the Cybersecurity and Infrastructure Security Agency] and this group [the Election Integrity Partnership]" and that emails show and EIP worked in concert with CISA to flag content for moderation. Hasan claimed this is a lie, that Taibbi conflated a private nonprofit with a government agency during his testimony, and that EIP only worked with the Center for Internet Security, or CIS, a private sector organization with a similar acronym, not CISA.  

It’s a serious charge and federal crime to make false statements to Congress, one punishable with prison time. 

But the record shows that CISA, the government agency, was involved in the very formation of EIP and was one of the most important government partners to the group in its bid to influence content moderation decisions at firms such as Facebook and Twitter. EIP’s own leaders have said as much, and there is endless documentation – from publicly available websites, to discovery from litigation, and from the “Twitter Files” – that all confirm this relationship and CISA’s role in assisting EIP.

In other words, Hasan is wrong and presents a deeply distorted view of the mechanics of this process. Taibbi’s testimony, especially the video excerpted by Hasan, is accurate. 

Worse, after being presented with evidence of CISA working directly with EIP, Hasan doubled down, repeating his false allegation that Taibbi lied under oath. In this post, I’ll provide a quick overview of the evidence, including new Twitter emails that have not been published before, that conclusively show that Taibbi was correct to connect CISA and EIP as partner organizations. Hasan’s repeated allegation is simply a bizarre and defamatory smear.

First, it’s worth understanding the acronyms underlying this story. CISA is the Cybersecurity and Infrastructure Security Agency, a sub-agency of the Department of Homeland Security, or DHS, that was created in 2018. The Election Integrity Partnership, or EIP, is an academic-led consortium that took shape in 2020 focused on election-related misinformation. And CIS is the Center for Internet Security, a nonprofit government contractor that manages information-sharing projects for CISA and DHS, among other clients.

I should disclose that I have a professional connection to this issue. I accompanied Taibbi once to Twitter headquarters in San Francisco, where I stopped by briefly to request and copy some emails. I used documents obtained from Twitter’s archives to do some reporting of my own. One story showed how the Pentagon operated a network of covert fake accounts to advance the U.S. military’s messaging operations in the Middle East. The documents also revealed that Twitter executives had full awareness of this program and refused to shut it down for years, despite promises to swiftly block state-backed influence operations. During my visit, Twitter had no input on my searches or reporting and to the best of my knowledge, the attorney conducting the document requests complied with whatever I asked. 

I’ve also been following the government agency CISA’s work for some time. As early as 2015, when the legislation enacting the agency was gaining steam, I wrote about how CISA represented a “major new privacy threat.” The legislation provided a special liability shield for corporations to “feed massive amounts of communications to private consortiums and the federal government” under the vague rubric of protection against undefined “cyber threats.”

Last year, in October, before the Twitter Files, and using primarily a DHS whistleblower and documents produced from litigation, Ken Klippenstein and I reported on the expansive role CISA had taken in seeking to shape political speech on social media platforms. The reporting charts the history of government interference in social media, from the early concerns around ISIS recruitment and 2016 Russian meddling, to the formation of the FBI's highly active task force that eventually corresponded with Twitter executives about content moderation issues on a near-daily basis. The reporting charted the evolution of CISA, which eventually decided to define election-relation communication on social media as a form of cyber security, as well as the agency’s close work with NGOs like EIP.

This history isn’t particularly hidden. EIP, in a publicly available report on its website, notes that the group was formed in consultation with CISA in July of 2020. The organization very clearly states that CISA is one of its main stakeholders, along with CIS, which was contracted to manage an information sharing consortium with CISA and local election officials called ISAC. Here’s the EIP report:

Alex Stamos, the director of EIP, gave a talk at the Atlantic Council, in which he outlined the "major stakeholders that we operated with, that we worked beside at EIP, our partners in government, most particularly those in CISA and DHS."

Internal Twitter emails that have not been published before, that I obtained in December of last year, show how this partnership functioned on a practical basis. On October 1, 2020, for example, then-Twitter attorney Stacia Cardille reached out to Yoel Roth, the former head of Trust and Safety at the company, to forward on an email chain from both CISA, the government agency, and CIS, the nonprofit with which it frequently works, regarding conservative tweets warning about the supposed dangers of vote-by-mail. 

The tweet in question was a claim that a California man had found unopened ballots in the dumpster, and a concern that this claim could represent misinformation that was being amplified by Russian sources. 

A member of EIP’s team and someone working at the California Secretary of State were the first people to flag the tweet, forwarding it onto the consortium led by CIS. From there, CIS and EIP, both nonprofits, categorized the tweet as a potentially dangerous form of misinformation that required action by social media companies, and emailed both CISA and Twitter to have the tweet taken down. 

The email chain between CIS and EIP officials and Twitter’s security team over just that one tweet show how closely these nonprofits work with the U.S. government’s national security arms. Officials at those organizations copied DHS official Brian Scully, who worked at CISA, along with other CISA email accounts, in their emails to Twitter executives calling for the company to remove the tweet. 

“Our hope is the platforms can do more to take down the misinformation. The EIP has been tracking this spread under ticket EIP-243 and has more examples,” Aaron Wilson, an election security official with CIS, wrote in an email to CISA, which was sent on to Twitter. 

Twitter executives discussed the issue, noting there was no evidence of foreign influence. But they nonetheless decided to delete the tweet. An archive of the tweet is here.

Another exchange between CISA and Twitter that I unearthed again illustrates how closely the nonprofits, EIP and CIS, work with the Department of Homeland Security. In that email thread, Robert Schaul, a DHS official at CISA, cited EIP’s internal flagging system when requesting that Twitter delete a tweet . 

There are two major issues presented by these exchanges. 

One of course is the content. Should the national security agencies like DHS and CISA interject into political speech issues? One could argue that they should as long as they are rigorously balanced, accurate and nonpartisan. Notably, Twitter executives censored Republican concerns around vote-by-mail, while allowing Democrats to spread conspiracy theories about the dangers of vote-by-mail. (For example, here’s an Eric Holder tweet raising baseless concerns against vote-by-mail, that was discussed internally at Twitter but not censored or fact-checked.) Still, this is a debate worth having that raises important constitutional and ethical questions. Many would say that there is no room for the government to be pressuring media outlets or social media firms on political speech grounds, especially around an election. 

The other issue is the one raised by Hasan and is simply functional. Did CISA play a role in EIP in its quest to shape content decisions at Twitter? The answer is obviously, “yes.”

Hasan claimed that only CIS, the government contractor, was involved, not CISA. That’s clearly not true, and not something that even CISA, EIP or CIS has claimed in the past. The government agency CISA worked hand-in-glove with CIS, in support of EIP. Hasan isn’t explaining any of this to his viewers and is instead engaged in a vicious character assassination effort. 

Where else can you find information on CISA working closely on election-related content moderation with the big social media firms? NBC News, the media outlet that is owned by Comcast, the cable giant that also employs Hasan, reported on this relationship during the election. If Hasan is still confused about the role of CISA, he could in fact ask his employer. Comcast is a partner to CISA through its consortium for communication platforms.

The Taibbi-Hasan debate speaks to the sorry state of affairs in the U.S. news media. Every journalist gets things wrong occasionally. Taibbi has conceded that he made an error in one of his tweets, though not in his congressional testimony, and swiftly corrected it. Many of Hasan’s claims have been debunked, including his false claim, first flagged by journalist Aaron Mate, that he “never said a word about the Hunter Biden story" and of course this CISA-EIP issue. Hasan’s version of journalism means never correcting his own falsehoods. But since Hasan works for a cable news network where exciting a polarized audience is the chief performance metric, he is sure to benefit from the gotcha-style assault on Taibbi.

The so-called debate is part of Hasan’s larger effort to obscure the importance of the revelations in the Twitter files by focusing obsessively on Elon Musk’s motives for allowing the disclosure of internal company documents. Hasan avoided, for example, addressing the substance of Taibbi’s reporting on the “Hamilton 68” dashboard. Taibbi’s reporting revealed that the tool, which Hasan’s employer MSNBC relied on liberally over the years, was essentially a fraud that relied on unfounded innuendo to smear critics of the national-security state as Russian stooges. Hasan, who has criticized national-security abuses in the past, has also failed to grapple with the Twitter Files’ exposure of the FBI’s persistent influence over Twitter’s content moderation policies. Hasan briefly discussed the latter topic, but never fully explained it to his MSNBC audience. Instead, we get a strange and personalized attack on Taibbi as a journalist, and a smear that he intentionally misled Congress under oath.

A brief note to readers:

Last month, I left The Intercept after more than eight years there. I’ll have more to say on that later. For now, I will be primarily focused on investigative reporting on this Substack. Look for much more original content here in the weeks and months to come. I’ll be breaking news on a range of topics. Please DM on Twitter with tips and ideas. I’m open to any thoughts.