Four years after the Twitter Files, the Missouri v. Biden case ends in a consent decree barring government from threatening protected speech - a belated but important victory
In the first week of December, 2022, a group of reporters now scattered and divided over the Iran War and other issues searched through a pile of raw correspondence at the San Francisco office of Twitter. One file we found was 67 pages of complaints about content, mostly from state officials, sent to an address marked misinformation@cissecurity. In one case, the Georgia Secretary of State’s office complained about a Fox5 Atlanta report titled, “Computer Problems Bring Down Voting Machines in Spaulding County.”
The story about technical difficulties in Spaulding County turned out to be accurate, as a “county-wide technical issue” delayed voting. Superior Court Judge Fletcher Sams ordered that ballots be kept open an extra two hours in 18 locations. In those same 67 pages we found a form letter informing the Georgia official that his complaint about the Fox story had already been forwarded to “our partners,” who included “The Cyber and Infrastructure Security Agency at the Department of Homeland Security” and “The Election Integrity Partnership” at Stanford University.
This was the first time an outsider had seen the plumbing of a wide-scale effort by federal agencies like the Cybersecurity and Infrastructure Security Agency (CISA) to regulate mis-, dis-, and malinformation in the social media landscape. It took considerable effort to untangle the mechanism by which complaints of “misinformation” were processed — the process was deliberately confusing — but the documents in the Twitter Files ended up playing a role in helping a landmark First Amendment case already launched in the courts, called Missouri v. Biden.
The general public, if it knows about this case at all, heard about the alleged “embarrassing defeat” plaintiffs like Dr. Aaron Kheriaty, future NIH Director Jay Bhattacharya, and Harvard’s Dr. Martin Kulldorff suffered in the Supreme Court in early 2024. This was often misreported, as the Supreme Court really punted on procedural issues and sent the case back down to be decided. It was forgotten until a consent decree was finally issued Tuesday, ending in victory for the plaintiffs. As Kheriaty wrote, the decree prohibits:
The U.S. Surgeon General, Centers for Disease Control and Prevention (CDC), and Cybersecurity and Infrastructure Security Agency (CISA) from threatening social media companies into removing or suppressing constitutionally protected speech on Facebook, Instagram, X (formerly Twitter), LinkedIn and YouTube.
“It was a victory, even if we did not get everything we wanted in the settlement,” said Kheriaty, the former professor of psychiatry at the University of California at Irvine, and chairman of the ethics committee at the California Department of State Hospitals.
Kheriaty, Bhattacharya, Hines, and others filed their first amended complaint on August 2, 2022 on the grounds that their social media posts about issues like Covid had been suppressed by a smorgasbord of government agencies, including the Department of Health and Human Services, the Centers for Disease Control and Prevention, and the Department of Homeland Security.
In that first complaint filed by the Missouri defendants, CISA was singled out based on evidence maladroitly conceded by Joe Biden’s own cabinet, as part of the argument for a controversial Disinformation Governance Board, which was to be headed by viral singing sensation Nina Jankowicz.
In a memo on September 13, 2021, DHS officials told then-Secretary Alejandro Mayorkas that “leading up to the 2020 election, CISA relayed reports of election disinformation from election officials to social media platform operators.” The Missouri plaintiffs didn’t know exactly what form this “relaying” took, but they correctly identified a serious problem: the government can’t pressure private carriers or publishers by labeling things “disinformation.”
This reflected a basic misunderstanding of the First Amendment, which was designed specifically to keep the state out of the Truth-Squadding business, particularly at moments like the Covid pandemic when the state might be seeking effective veto power over critical national issues.
The Consent Decree handed down this week put into the books concepts that should provide a little comfort to those of us who fought to surface this issue. One principle agreed upon is that “modern technology does not alter the government’s obligation to abide by the strictures of the First Amendment.” It also enjoined government agencies from threatening social media companies with “some sort of punishment” if they don’t “remove, delete, suppress, or reduce, including through altering their algorithms, posted social media content containing protected speech.”
Kheriaty talked about how he understood the accomplishments of Missouri v. Biden. His thoughts are similar to how I feel about the Twitter Files. The court case and the Twitter documents took a matter that was not only a non-issue in media (the existence of censorship programs was routinely denied before late 2022) and rapidly propelled it to the center of the American political discussion.
It reached the Supreme Court and, as Aaron pointed out, ended up being an issue that helped decide the 2024 election. In the Vice Presidential debate in 2022, when J.D. Vance was asked if his running mate was a threat to democracy, he responded that the widespread assault on the First Amendment was the threat he worried about:
In the intervening years, I’ve had to think about this a lot. I had to think about it last week. I had to think about it yesterday, and again today. Friends and colleagues regularly challenge the utility of a court case and scandal that allowed Trump and his own more-than-questionable approach to speech issues a chance to prevail in 2024, by capitalizing on Joe Biden’s idiotic government-wide jawboning program.
To this I ask, what was the alternative? Letting it go? A ruling permitting the behaviors detailed in Missouri v. Biden would have been far more devastating. If you’re concerned about a hyper-empowered chief executive intent on deamplifying, say, derogatory content about the war in Iran, you need it enshrined in law that threats and pressure to social media companies are strictly forbidden. In that regard, everyone irrespective of party should be happy about this result.
As for the 2024 election, Democrats had plenty of time to get on the right side of these issues. I know this because I privately pleaded with plenty of them behind the scenes to see how badly this was going to play in an election year. They not only refused to listen, but insisted on nominating two hardcore speech ignoramuses in Kamala Harris and Tim Walz, while smearing people like Kheriaty for bringing this case.
After the many controversies of the last few years, Democrats can probably be counted on to support the First Amendment for as long as Trump is in office, and you know what? That’s okay! That’s good. As long as both parties see the political benefit in appearing to be on the side of one of America’s core beliefs, it’s a win. It’s a testament to who really deserves the credit for decisions like this Decree: the public. Enough people expressed enough disgust about these behaviors that the First Amendment has been updated in the books, boasting a fresh coat of paint for the social media age. It’s good for everyone. When was the last time we could say that?
Congrats to Aaron and his co-plaintiffs, who went through a lot on the road to this result. Historians won’t know what a disgusting process it was to get here, but I’ll remember, and I hope Racket readers will as well. The plaintiffs who hung in deserve a hearty pat on the back. As John Vecchione, counsel for the New Civil Liberties Alliance put it, “Freedom of speech has been powerfully preserved by our clients.” It’s true, and a happy thing that a few people cared enough to see it through.
Great news Matt!! Thank you so much for carrying the water and keeping a light on this! Really happy for the plantiffs as well!
Cheers to all lovers of free speech!
Taibbi, you did so much, along with Michael Shellenberger and others - this is a win you influenced mightily.
Facts Only
In December 2022, reporters examined Twitter’s internal files, uncovering 67 pages of complaints about content, primarily from state officials, sent to a misinformation reporting address linked to CISA and Stanford’s Election Integrity Partnership.
A complaint from the Georgia Secretary of State’s office targeted a Fox5 Atlanta report on voting machine issues in Spaulding County, which was later confirmed accurate by a court order extending voting hours.
The *Twitter Files* revealed federal agencies, including CISA, were involved in relaying "misinformation" complaints to social media platforms, a process later scrutinized in *Missouri v. Biden*.
Plaintiffs Dr. Aaron Kheriaty, Dr. Jay Bhattacharya, and Dr. Martin Kulldorff filed their first amended complaint on August 2, 2022, alleging their COVID-19-related social media posts were suppressed due to government pressure.
A September 2021 DHS memo acknowledged CISA’s role in relaying election disinformation reports to social media platforms ahead of the 2020 election.
The Supreme Court initially remanded *Missouri v. Biden* in early 2024 on procedural grounds before the case returned to lower courts.
A consent decree issued in July 2024 prohibits the U.S. Surgeon General, CDC, and CISA from threatening social media companies to remove or suppress protected speech on platforms like Facebook, X (Twitter), and YouTube.
The decree states that "modern technology does not alter the government’s obligation to abide by the First Amendment."
Plaintiffs argued that government agencies pressured platforms to censor speech by labeling it "disinformation," violating First Amendment protections.
The case gained political traction, with figures like J.D. Vance citing government censorship as a threat to democracy during the 2022 vice-presidential debate.
The decree was celebrated by free speech advocates but criticized by those concerned about its potential impact on combating misinformation.
The legal battle lasted over two years, with plaintiffs facing significant pushback before securing the decree.
Executive Summary
A landmark First Amendment case, *Missouri v. Biden*, concluded with a consent decree barring federal agencies from pressuring social media companies to suppress constitutionally protected speech. The case originated from evidence uncovered in the *Twitter Files*, which revealed coordinated efforts by agencies like CISA, the CDC, and the Surgeon General’s office to flag and remove content deemed "misinformation," particularly during the 2020 election and the COVID-19 pandemic. Plaintiffs, including Dr. Aaron Kheriaty and Dr. Jay Bhattacharya, argued that their social media posts about COVID-19 were unjustly censored due to government coercion. While the Supreme Court initially remanded the case on procedural grounds, the final decree prohibits agencies from threatening platforms to remove protected speech, affirming that modern technology does not alter First Amendment obligations.
The case highlights a broader debate about government overreach in digital speech regulation. Critics of the decree argue that it may hinder efforts to combat genuine misinformation, while supporters see it as a necessary check on state power. The ruling arrives amid ongoing political divisions, with some framing it as a victory for free speech and others warning of potential consequences for public health communication. The decree’s language explicitly forbids agencies from leveraging "punishment" to influence content moderation, marking a significant legal precedent for the social media era.
Full Take
**Steelman:** The *Missouri v. Biden* consent decree represents a critical victory for free speech advocates, codifying protections against government coercion of social media platforms. The evidence from the *Twitter Files* and internal memos demonstrated a systemic effort by federal agencies to influence content moderation under the guise of combating misinformation. The decree’s language is unambiguous: agencies cannot threaten platforms to suppress protected speech, even in contexts like public health or elections. This sets a vital precedent, ensuring that the First Amendment’s safeguards extend into the digital age. The plaintiffs’ persistence—despite political and legal hurdles—highlights the importance of holding institutions accountable for overreach.
**Pattern Scan:** The narrative leans into a classic *ARC-0024 Ambiguity* frame, where "misinformation" is treated as a monolithic threat rather than a contested category. The article also employs *ARC-0043 Motte-and-Bailey*, oscillating between broad claims of government censorship and specific examples (e.g., the Georgia voting machine story) that, while valid, may not fully represent the scope of the issue. The framing of the decree as an unalloyed win for free speech risks *ARC-0012 False Binary*, ignoring legitimate concerns about how platforms might now handle harmful but legally protected content (e.g., medical misinformation during a pandemic). The emotional tone—celebrating the plaintiffs’ victory while dismissing counterarguments as "idiotic" or "ignoramus"—hints at *ARC-0003 Weaponized Anger*, though it stops short of outright rage bait.
**Root Cause:** The underlying paradigm is a clash between two competing visions of governance in the digital public square: one that prioritizes absolute free speech (even at the cost of societal harm) and another that seeks to balance speech with collective safety. The decree reflects a libertarian-leaning interpretation of the First Amendment, where government influence—direct or indirect—is treated as inherently corrupting. This echoes historical tensions, such as the Red Scare’s censorship or the Pentagon Papers case, but transposes them onto a decentralized, algorithm-driven media landscape. The unstated assumption is that private platforms, freed from government pressure, will self-regulate fairly—a proposition that remains untested.
**Implications:** For human agency, the decree empowers individuals to challenge government-overseen speech suppression, but it also shifts power to unaccountable platform algorithms. The costs may fall disproportionately on marginalized voices, whose speech is often the first to be labeled "misinformation" in polarized debates. Second-order effects could include a chilling effect on government public health communication (e.g., during future pandemics) or a surge in unmoderated harmful content, forcing platforms to either over-censor or abandon moderation entirely. The political fallout is already visible, with both parties weaponizing the ruling—Democrats now feigning free speech advocacy to oppose Trump, Republicans using it to undermine trust in institutions.
**Bridge Questions:**
If government pressure is prohibited, what mechanisms can ensure platforms don’t suppress speech arbitrarily in the name of "community standards"?
How might this decree interact with existing laws (e.g., Section 230) that already shape platform liability and moderation?
What safeguards could prevent bad actors from exploiting this ruling to amplify genuinely harmful content (e.g., election interference, health hoaxes)?
**Counterstrike Scan:** A coordinated influence campaign pushing this narrative would amplify the "government censorship" frame while downplaying the complexities of misinformation’s real-world harms. It would leverage emotional triggers (e.g., "belated victory," "disgusting process") to rally support, frame opponents as authoritarian, and present the decree as a binary choice between freedom and tyranny. The actual content aligns partially with this playbook—particularly in its dismissive tone toward critics—but stops short of full manipulation. It acknowledges counterarguments (e.g., Democratic shifts on free speech) and provides concrete legal details, which undermines a pure propaganda structure. The focus on procedural wins over systemic analysis suggests a strategic emphasis on mobilizing a specific audience rather than fostering broad consensus.
**Patterns detected:** *ARC-0024 Ambiguity, ARC-0043 Motte-and-Bailey, ARC-0012 False Binary, ARC-0003 Weaponized Anger*
Sentinel — Human
The text shows signs consistent with human authorship. The author's personal voice, inconsistent sentence lengths, and lack of clear coordination indicators suggest that this is likely a human-written article.