Luján, Colleagues Urge Social Media and Encrypted Messages Companies to Increase Resources Toward Combatting 2024 U.S. Election Disinformation

Washington, D.C. – U.S. Senator Ben Ray Luján (D-N.M.) led a letter signed by U.S. Senators Jeanne Shaheen (D-N.H.), Elizabeth Warren (D-Mass.), Ron Wyden (D-Ore.), and Jeff Merkley (D-Ore.) to 11 of the largest and most popular social media and encrypted chat companies in the U.S. calling for increased resources toward combatting 2024 U.S. election administration and certification disinformation. 

“We are deeply concerned that the dissemination of election disinformation via your products and/or platforms – if left unmitigated – will suppress voter participation, sow doubt in U.S. democracy and incite political violence,” wrote the Senators. “Considering the increase in election disinformation on digital platforms during recent elections, there is ample cause for concern.”

“During the 2020 and 2022 U.S. federal elections, foreign adversaries supported the creation and targeting of election disinformation to undermine our democracy. During the 2020 elections, research showed that election disinformation in Spanish stayed up for longer on social media, as compared to English,” the Senators continued.

The Senators urged Meta, Google (YouTube), TikTok, X (Twitter), Reddit, Snapchat, Amazon (Twitch), Discord, Signal, Telegram and Apple (Messages) to:

• Share information about the size and capacity of their 2024 U.S. elections safety resourcing – including personnel and technologies – broken down by language;

• Commit to increasing their 2024 U.S. election safety team and technology resourcing for the 10 most commonly spoken languages on their platform(s); 
• Share information about how they plan to de-amplify and/or remove election disinformation (whether created using AI or not) and/or user accounts who spread this disinformation, when in violation of their policies;
• Share their plans for amplifying official election information before, during and after the 2024 U.S. elections; the letter also encourages companies to offer translation of official election information as a public service; 
• For encrypted chat providers, explain whether they have a reporting system for their users to flag unwanted election disinformation and what enforcement measures are in place. 

The full text of the letter is available HERE and below: 

We write to you concerning the potential spread of election administration and certification disinformation before, during and after the 2024 United States elections. We are deeply concerned that the dissemination of election disinformation via your product(s) and/or platform(s) – if left unmitigated – will suppress voter participation, sow doubt in U.S. election processes and incite political violence. 

For the purposes of this letter, “election disinformation” refers to false content about 2024 U.S. elections administration or certification – including falsehoods that pertain to the time, place and manner of the elections, voter eligibility requirements, and the processes of ballot counting and result certification – that is intentionally created to mislead voters and/or to suppress voter turnout.¹ 

Considering the increase in election disinformation on digital platforms during recent elections, there is ample cause for concern. During the 2020 and 2022 U.S. federal elections, foreign adversaries supported the creation and targeting of election disinformation to undermine our democracy.² During the 2020 elections, research showed that election disinformation in Spanish stayed up for longer on social media, as compared to English³; and recent research from the University of Washington points to how Spanish-English language disparities in the availability and persistence of election disinformation continue.⁴ In 2022, the National Association of Latino Elected and Appointed Officials (NALEO) found that 62% of Spanish-speaking Latino households had “seen or heard a lot” about the “Big Lie” or false claims about U.S. elections administration and certification being fraudulent – 22% more than English-speaking Latino households.⁵ Meanwhile, a former U.S. President has spread and continues to spread dangerous election disinformation, despite no evidence supporting these claims.⁶ 

Recent advances in artificial intelligence (AI) technologies have only strengthened foreign and domestic malign actors’ capability to create, micro-target and spread election disinformation quickly and on a large scale. Just last week, the U.S. Department of Justice seized 32 Russian-operated internet domains that covertly promoted AI-generated false narratives on social media to target specific American demographics and regions in an effort to subvert the 2024 elections.⁷ Notably, as a 2024 report from the Microsoft Threat Analysis Center documents, although election influence operations from China, Iran and Russia have started ramping up, their tactics are not limited to AI: “we’ve seen nearly all actors seek to incorporate AI content in their operations, but more recently many actors have pivoted back to techniques that have proven effective in the past—simple digital manipulations, mischaracterization of content, and use of trusted labels or logos atop false information.”⁸ 

We ask that you provide a response to the following questions by October 3, 2024. Unless specified, all questions below pertain to the U.S. 2024 elections and use the definition of ‘election disinformation’ discussed earlier in this letter. 

Questions:

Enforcement of policies pertaining to election disinformation:

• What are your company’s current policies pertaining to election disinformation, and what are your enforcement processes for these policies? 
• In what situations is election-related content – whether flagged through internal processes or because of user input – referred for human review? What is the average timeline for completing such a review? 
• What are your policies on warning and/or banning a user account after a certain number of violative content-sends or posts?
• What technical interventions do you have in place to limit the virality of election disinformation?
• Do you temporarily limit the virality of election-related content that has been flagged and is waiting for human review?
• What is the mean and median time from the reporting of violating content and/or users to when violating content/users are removed, de-amplified and/or downranked? Please break down these metrics by language, for the last year.

Resourcing for election safety teams:

• Which teams at your company focus on enforcing your policies regarding U.S. election disinformation, and for which product(s)/platform(s)? Describe each team’s functions, goals and responsibilities, and for how long these personnel will focus specifically on the 2024 U.S. elections.
• How many full-time and part-time employees and contractors work on the team(s)? 
• Please provide a breakdown of the number of personnel focused on 2024 U.S. election integrity per content-language and explain any differences in resourcing across languages.
• Has your U.S. election safety team(s) size and composition changed over the last five years, and if so, how? Please share decreases and increases in headcount per role or team, per year, and explain any changes.

De-amplification commitment: 

• For the 2024 U.S. elections, after a full adjudication per your platform policies, how quickly can your company commit to taking action on content deemed to be election disinformation and/or user accounts deemed to be spreading election disinformation? 
• How will you apply these policies equally across all user accounts?

Equitable disinformation mitigation across languages:

• For the 2024 U.S. elections, will your company commit to providing equally effective election disinformation mitigation across the 10 most commonly spoken languages on your product(s)/platform(s) in the U.S.?

Increase election safety resourcing before, during and after the 2024 elections:

• Will your company commit to increasing personnel and technical resources to enforce your election integrity policies from now until November 5th, 2024, and during the election certification period between November 5th and January 20th?
• If yes, please outline the additional steps that your company will commit to taking, the start and end-dates of these additional steps, and how you will measure and report the success of these steps.

Amplifying official election information sources:

• What is your company’s plan to amplify (for example, promote, link to and/or encourage the sharing of) official election administration and certification information pertaining to the 2024 U.S. elections?
• Will your company commit to measuring the reach and engagement that results from these amplification efforts?
• Is translation of official election information into the 10 most commonly spoken languages on your product(s) and/or platform(s) a step that your company has the capacity to and interest in offering as a public service during the 2024 U.S. elections?

Account-level mitigation strategies for one-to-one and group chat service providers, including encrypted chat providers: 

• Does your company offer a process for your customers to flag unsolicited election disinformation that was sent to them by a specific user account? Explain your policies towards the spread of unwanted election disinformation on your chat product(s) and/or platform(s), and how your teams enforce these policies without infringing upon user privacy and/or encrypted communication services.
• What are your policies on warning and/or banning a user account after that account sends a certain number of messages that violate your policies?

Thank you for your prompt attention to these questions and calls for commitment. We look forward to hearing from you. 

###