Partnership targeted election misinformation, not conservatives

A former congressional candidate claims that he and other conservatives were targeted on social media with the government’s help around the time of the 2020 election. 

"(The) Department of Homeland Security and the State Department coordinated with a group called the ‘Election Integrity Partnership’ to censor Joe Biden’s opponents," says the caption on an Oct. 5 Instagram post by Robby Starbuck. 

Starbuck is a Republican who announced in November 2020 that he would run for Congress in Tennessee before his own party kicked him off the primary ballot in April 2022. He later filed to be a write-in candidate for the August election after losing his challenge to stay on the ballot.

Starbuck’s caption also says the "Biden regime paid ‘Election Integrity Partnership’ entities $12+ MILLION from taxpayer funds" and that Starbuck was added to the partnership’s "enemies list." He detailed his claims in an accompanying 50-minute video, as well as in a lengthy thread on Twitter. 

The Election Integrity Partnership, which conducted its work publicly throughout 2020 and published a final report of its findings in March 2021, has drawn criticism in conservative media outlets and seen online attacks against its researchers in recent weeks.

The Instagram post was flagged as part of Facebook’s efforts to combat false news and misinformation on its News Feed. (Read more about our partnership with Meta, which owns Facebook and Instagram.)

The Department of Homeland Security’s Cybersecurity and Infrastructure Security Agency and the State Department’s Global Engagement Center were among several organizations that worked with the Election Integrity Partnership — a collaboration of researchers — on 2020 election misinformation claims. The working relationship between the federal agencies and the Election Integrity Partnership began during the tenure of former President Donald Trump.

In practice, though, no one from CISA sent examples of misinformation to the Election Integrity Partnership, and the partnership sent no reports to social media companies on the agency’s behalf, a blog post from the partnership stated.

The State Department’s Global Engagement Center, which focuses on foreign propaganda and disinformation, submitted some misinformation claims to the partnership, though it’s unclear how many. However, less than 1% of the claims investigated by the partnership focused on claims involving foreign actors, the report said.

There’s no evidence the partnership or government agencies working with it sought to "censor" conservatives.

Groups involved in the partnership have received more than $12 million in federal grants since Biden took office, and $3 million of those grants were related to the partnership's work.

There is no evidence that the grants were politically motivated. Starbuck told PolitiFact he didn’t say the grants were connected to the partnership’s work in 2020, but "people can decide for themselves what they think about it." 

The partnership, which began in July 2020, has a goal to work with government and civilian stakeholders to combat mis- and disinformation on social media about voting and elections before and after the hotly contested 2020 presidential election. The partnership said in July 2022 that it would continue its work in this year’s midterm elections.

Four research groups collaborate on the project: the University of Washington’s Center for an Informed Public; the Stanford Internet Observatory; Graphika, a social analytics firm; and the Atlantic Council’s Digital Forensic Research Lab.

The partnership said when announcing its creation that its mission was to support "real-time information exchange between the research community, election officials, government agencies, civil society organizations, and social media platforms." It was focused narrowly on content "intended to suppress voting, reduce participation, confuse voters as to election processes, or delegitimize election results without evidence." 

The partnership laid out how it conducted its work, what it found and its recommendations to safeguard future elections in a 290-page report titled "The Long Fuse: Misinformation and the 2020 Election," which it published in March 2021. 

The partnership initially was set up in consultation with DHS’ Cybersecurity and Infrastructure Security Agency. The idea for the partnership began with four Stanford Internet Observatory students who were interning at CISA, which is focused on protecting against cyber threats and working on election infrastructure security, and often partners with the private sector.

On Oct. 5, in response to recent attacks on its work, the Election Integrity Partnership said in a blog post that its partnership with CISA was reviewed and approved by Trump administration attorneys. The partnership’s aim was helping CISA "understand rumors and disinformation around the 2020 election and so CISA could provide corrective and/or clarifying information from election officials," it said.

Kim Wyman, a senior election security advisor at CISA, rejected claims of censorship.

"CISA does not censor speech, period," Wyman said in a statement to PolitiFact, adding that the agency works "in a non-partisan manner with state and local election officials" to provide voters with accurate information about elections. "Online content platform operators make their own decisions."

Renée DiResta, a technical research manager at the Stanford Internet Observatory, explained how the Election Integrity Partnership was formed in this video from a 2021 cybersecurity summit. (CISA)

The partnership allowed its own researchers and what it deemed "trusted" external stakeholders to flag social media content or communicate with analysts regarding online claims.

Featured Fact-check

Facebook posts

stated on March 24, 2024 in a Facebook post

Donald Trump won Pennsylvania in the 2020 presidential election.

The external stakeholders listed in its March 2021 report included government agencies such as CISA and the State Department’s Global Engagement Center.

Also listed was the Election Infrastructure Information Sharing and Analysis Center, which is a elections and cybersecurity organization that’s a division of the nonprofit Center for Information Security. It gets some of its funding through the Department of Homeland Security. 

Civil organizations, such as the NAACP, AARP and others, also could flag content and communicate with analysts. The major social platforms — Facebook and Instagram, Google, YouTube, Twitter, TikTok, Reddit, Nextdoor, Discord and Pinterest —  also signed on to the partnership. 

Three teams evaluated flagged content. 

• The first team determined whether the concern was within the project’s scope and checked the claim’s veracity; 

• The second team delved deeper into the claim and recommended action, such as communicating to partners if appropriate; 

• A third team of managers from the four research groups could tag platform partners on a claim, communicate with government partners to get information, or close a case.

Overall, the partnership said it processed 639 claims that it considered to be within the project’s scope, 72% of which were related to "delegitimizing the election results." 

The partnership did not recommend specific actions to social media platforms, but tagged them on reports if researchers believed the claims violated the platforms’ published policies. Those platforms "made their own determinations on how to treat our reports," the partnership said.

It noted that 35% of the links shared with Facebook, Instagram, Twitter, TikTok and YouTube were either "labeled, removed, or soft-blocked." (PolitiFact has a partnership with Meta, which owns Facebook and Instagram, and you can read more about it here.)

About 79% of the claims analyzed were submitted by analysts from the four research organizations of the partnership, rather than external partners. An additional 16% were reported by the nonprofit Center for Internet Security as tips from election officials.

That accounts for about 95% of claims submitted overall. Other stakeholders submitting claims were MITRE, Common Cause, the Democratic National Committee, the Defending Digital Democracy Project and the NAACP, the report said. 

Both the Democratic and Republican national committees were invited to submit tickets. The DNC submitted four reports. The RNC did not respond to the partnership’s outreach and submitted no reports.

Starbuck claimed in his Instagram post that he was named in an "enemies list" in the Election Integrity Partnership’s March 2021 report, along with other conservative figures, such as Donald Trump Jr.; U.S. Rep. Marjorie Taylor Greene, R-Ga.; and Charlie Kirk, founder and president of Turning Point USA, a politically conservative group advocating for limited government.

Starbuck is referring to Sec. 5.5 in the report, which includes a list of what it called "the 21 most prominent repeat spreaders on Twitter." The list is based on Twitter data collected through Dec. 12, 2020, approximately one month after Starbuck announced his run for Congress

That was the only mention of Starbuck in the report. Although the "repeat spreaders" section listed only conservative Twitter accounts, other parts of the report mentioned left-leaning accounts. For example, Occupy Democrats was included for posts that "functioned to fan fears of voter disenfranchisement and intimidation."

Overall, the report concluded that while "false narratives occasionally gained traction on the political left, almost all of the most prominent repeat spreaders …  belonged to conservative and/or pro-Trump individuals and organizations." 

The partnership said its lists of repeat spreaders, although published in its final report and in blog posts, "were not provided directly to any partner organization," such as Twitter. 

Starbuck claimed that the Department of Homeland Security and State Department coordinated with the Election Integrity Partnership to "censor Joe Biden’s opponents."

Agencies within those departments were external stakeholders in the Election Integrity Partnership during its 2020 work, before Biden took office. 

CISA did not pass along any claims to the partnership, which also said it didn’t forward any reports on CISA’s behalf to social media platforms. 

It’s unclear how many misinformation claims were submitted to the partnership by the State Department’s Global Engagement Center, which focuses on foreign propaganda and disinformation efforts. But less than 1% of the claims investigated by the partnership focused on claims involving foreign actors.

Although both government agencies were involved in the partnership, there is no evidence that either of them, or the partnership’s researchers, acted to "censor Joe Biden’s opponents." 

We rate this claim False.

Updated, Oct. 13, 2022: The story has been updated to include a statement from CISA that was received after publication.