Judicial Watch: Homeland Security Records Confirm Extensive Government Censorship of Conservative Social Media During 2020 Election

(Washington, DC) – Judicial Watch announced today that it received 274 pages of records from the U.S. Department of Homeland Security (DHS) revealing an extensive effort by government and non-government entities to monitor and censor social media posts on fraud during the 2020 election.

The records were uncovered thanks to a Judicial Watch November 2022 Freedom of Information Act (FOIA) lawsuit for communications between the Cybersecurity and Infrastructure Security Agency (CISA), a division of DHS, and the Election Integrity Partnership (EIP), which was created to flag online election content for censorship and suppression (Judicial Watch Inc. vs. U.S. Department of Homeland Security (No. 1:22-cv-03560 )).  

A message on September 29, 2020, at from “[email protected]” to “Misinformation Reports” at CIS (Center for Internet Security, a CISA-funded non-profit) has the subject line: “EIP-257Case #CIS – MIS000019: CT [Connecticut] Misinformation related to absentee ballots and fraud” and states:

Analysts at the EIP [Election Integrity Partnership] looked for cross-platform spread of this post, including in local media, but it appears contained to Facebook at this point.

The issue has been raised with Facebook to see if additional steps can be taken on the platform.

Included in the chain from [redacted] to “Misinformation Reports” is a reply: 

Hey [redacted]

The platforms may or may not take action, and they sometimes take action without notifying EIP [Election Integrity Partnership]. “Done” is a signal to other EIP analysts that no more analysis needs to be done on their end at this time.

What we can be sure about at this stage is that Facebook have been notified, but nothing else it seems. EIP will periodically re-check the link to see if the content has been removed and will update the ticket if it has. Facebook may also be investigating the misinformation and may update the ticket itself if an action is taken. 

A response from a person at Stanford whose name is redacted reads:

Just wanted to add in, it seems like the links are now down. I agree that this should not have moved into Done without sending a summary to the reporting partners (you all) and I’ve talked to the team about this, we will have that for the next one! Facebook doesn’t tell us specifically when they take action sometimes so we just have to keep checking the links. I’ve asked the team to go write a summary for this incident, let us know if we can provide further help here.

 Also, on September 29, 2020, “[email protected]” writes to [redacted] regarding Connecticut absentee ballots:

Thanks for flagging this. Our analysts looked into the claim – the specific post was spread through a personal network but didn’t take off further across social media platforms. We flagged the post to Facebook for removal and the link is no longer active which means it has either been taken down or made private to the individual’s Facebook. We are monitoring similar narratives in case this kind of message comes up again

An October 1, 2020, message from a person at CIS using the “[email protected]” address responds:

Thanks EIP. We have kept the election official updated on the steps you took. We also received confirmation from Facebook (by way of CISA) that Facebook took action on this case. 

The records include a September 26, 2020, email from Trevor Timmons, chief information officer of the Colorado Secretary of State, to Central Cyber [redacted] @cisecurity.org, using the subject line “Reporting Twitter post with misinformation on recent Colorado mailing:”

We’d like Twitter to review recent posts from a Denver news channel that contain misinformation about a recent mailing from the Colorado Secretary of State’s office.

[Redacted Twitter posts]

The posts have been highlighted as containing false information by personnel from our office, other media sources, and others. At best, we’d suggest they be removed as promoting inaccurate and false information. At a minimum, we’d request they be labeled as “false.”

Brian Scully, who was head of the Mis-, Dis-, Malinformation (MDM) branch at the National Risk Management Center at DHS, forwards this to an individual whose name and email address are redacted, commenting: “This came in from Colorado over the weekend. Twitter did not take either down.” 

On October 1, 2020, “[email protected]” writes to someone whose name is redacted using the subject line “Fraudulent Facebook Site Hood River County, OR:”

I’m adding Facebook to this ticket, and recommending that they take action on this page.

We believe that this page was automatically generated after someone running for Mayor in Hood River County repeatedly tagged themselves at the “Hudson River County Elections Office”. This person uses two Facebook accounts with different names [redacted]  which is why two people have checked into this location. Their repeated check-ins triggered the creation of this page, which now exclusively features their location-–tagged posts with their commentary about local politics.

Our understanding is that another verified page, such as the Hood River County page, could claim or request deletion for this page. However, we recommend that Facebook take immediate action on this page (and similar unofficial pages for election-related organizations), due to its potential for spreading misinformation about voting processes in Hood River County.

On the same day “[email protected]” comments on the same string:

Additionally, Facebook should more generally review exploitation of the “unofficial pages” feature in the context of elections. Can page creation see a lag or delay in order to prevent brigading in the final weeks of the election?

In this case, the tagging activity began in May 2019, initially by a mayoral candidate, then by others who have turned it into an unofficial town forum. It may not always be benign.

On October 5, 2020, a person whose name is redacted writes to Matthew Masterson [formerly with DHS and Stanford Internet Observatory], and Brian Scully of the DHS, plus numerous other redacted individuals, using the subject line “EIP-CIS Sync:”

Hi all,

The [email protected] reporting system is now up and running, as is EIP’s inbound and outbound tip system. This call is to discuss how this process has gone so far, and to nail down the EIP <> ISAC SLA moving forward.

On October 27, 2020, Scully writes to “[email protected]” and other redacted individuals, using the subject line “Duplicate reporting to Twitter:”

Twitter asked if we could have reports coming through misinfo reports only be reported to them via CISA. I know the EIP is sharing some of the reports with them as well and they are trying to cut back on duplicates. Any issues with this approach? From a process standpoint it would look like – election official> [misinformationreport]s > CISA/EIP/Others > CISA sends to Twitter> rest of process is the same. 

On October 28, 2020, someone whose name is redacted responds:

Hello Brian, 

Sorry for the delayed response here – took a day to settle this out on the EIP side. We are doing our best effort on the CIS reports to not duplicate notifications to Twitter when its (a) marked as Twitter being notified (assumption is everything sent with CIS #XXXXX has made it to Twitter) and (b) we have no substantial research to add to the matter. There have been a couple cases in the last week where we did not escalate CIS reports due to this, and we will continue to push on this front. Given how close we are to the election, I think this is the best option so that every report goes through CISA to the platforms, and EIP also sends over the relevant information only when we have further analysis to add which might be of help (more links, evidence of origination or coordination, etc.). We don’t want to overwhelm them, but on the day of, I think we’d kick ourselves more for not alerting than for double reporting.

Does this sound okay? We have communicated this to our Twitter partners through our channel with them. 

Scully responds in part: 

I’m not sure how we’d go from double reporting to no reporting in a scenario where a report comes through CIS as CISA will always report those to Twitter, but agree more reporting is better than less.

On November 2, 2020, Scully writes to a redacted recipient:

Is EIP open to receiving reports from civil society groups? Was asked by our friends at Harvard. 

A person whose name is withheld responds:

Hey Brian,

EIP takes tips from many civil society partners – I’m sure they’d welcome intake from Harvard. The same tip address should be fine. 

On November 8, 2020, a message from “[email protected]” is sent to individuals whose names have been redacted, using the subject line “Antrim County, MI election results error in reporting:”

Hello ISAC [Information Sharing and Analysis Center, a CISA-supported collaboration]– We are moving you to https://2020partnership.atlassian.net/servicedesk/customer/portal/5/EIP-949 which Facebook, Twitter, and Google are also on. We tweeted about this claim last night (here), and continued to track the growth of this narrative through today. 

This narrative, which started around the situation in Antrim, has now been used to throw doubts on the results in Georgia and other regions.

On November 9, 2020, someone at CIS using “[email protected]” writes to people whose names are withheld, using the subject line: “Million MAGA March – Nov 14th – Telegram:”

Thanks. I have spoken to [redacted] and this to make sure they are keeping the right authorities in sync.

On November 9, 2020, “[email protected]” sends a message to individuals whose names are redacted, using the subject line “Nevada GOP claims in Twitter thread they found ballots that could have been filled out by anyone”

Thanks, Twitter has received and is reviewing.

A November 10, 2020, message from “[email protected]” to persons whose names are withheld, using the subject line ‘Million MAGA March – Nov 14th – Telegram,” includes a chart showing the number of tweets on the subject and the message:

ISAC, please find an updated graph showing viral spread of this event on Twitter. Growth is linear and does not yet show signs of diminishing. We will continue to monitor.

“These new, previously secret documents show that the Deep State worked full time to censor and monitor Americans in collusion with left-wing interest groups and Big Tech,” Judicial Watch President Tom Fitton said. “The Trump administration should launch a criminal probe into this assault on the First Amendment civil rights of Americans.”

In August 2023, Judicial Watch filed two FOIA lawsuits against the U.S. Department of Justice and other federal agencies for communications between the agencies and Facebook and Twitter regarding the government’s involvement in content moderation and censorship on the social media platforms. 

In June 2023, Judicial Watch sued DHS for all records of communications tied to the Election Integrity Partnership. Based on representations from the EIP (see here and here), the federal government, social media companies, the EIP, the Center for Internet Security (a non-profit organization funded partly by DHS and the Defense Department) and numerous other leftist groups communicated privately via the Jira software platform developed by Atlassian.

In February 2023, Judicial Watch sued the U.S. Department Homeland Security (DHS) for records showing cooperation between the Cybersecurity and Information Security Agency (CISA) and social media platforms to censor and suppress free speech.

Judicial Watch in January 2023 sued the DOJ for records of communications between the Federal Bureau of Investigation (FBI) and social media sites regarding foreign influence in elections, as well as the Hunter Biden laptop story.

In September 2022, Judicial Watch sued the Secretary of State of the State of California for having YouTube censor a Judicial Watch election integrity video.

In May 2022, YouTube censored a Judicial Watch video about Biden corruption and election integrity issues in the 2020 election. The video, titled “Impeach? Biden Corruption Threatens National Security,” was falsely determined to be “election misinformation” and removed by YouTube, and Judicial Watch’s YouTube account was suspended for a week. The video featured an interview of Judicial Watch President Tom Fitton. Judicial Watch continues to post its video content on its Rumble channel (https://rumble.com/vz7aof-fitton-impeach-biden-corruption-threatens-national-security.html).

In July 2021, Judicial Watch uncovered records from the Centers for Disease Control and Prevention (CDC), which revealed that Facebook coordinated closely with the CDC to control the COVID narrative and “misinformation” and that over $3.5 million in free advertising given to the CDC by social media companies.

In May 2021, Judicial Watch revealed documents showing that Iowa state officials pressured social media companies Twitter and Facebook to censor posts about the 2020 election. 

In April 2021, Judicial Watch published documents revealing how California state officials pressured social media companies (Twitter, Facebook, Google (YouTube)) to censor posts about the 2020 election.

###