Don’t Mourn the Fact-Checkers
Misguided decisions and the superiority of crowdsourcing mean Zuckerberg was right to change direction. But he must also keep Trump at arm’s length.
The Great American Vibe Shift, rapidly reshaping U.S. culture and politics, has landed at 1 Hacker Way—the gleaming Palo Alto headquarters of Meta. In a video this week, Meta founder and CEO Mark Zuckerberg announced sweeping changes to content policies on Facebook, Instagram, and Threads to enhance free expression and reduce censorship. Key changes include phasing out third-party fact-checking in the United States in favor of crowdsourced fact-checking, similar to X’s Community Notes feature, which allows users to add context to misleading posts. Zuckerberg also pledged to loosen restrictions on “hate speech” and to limit automated filtering to “illegal and high-severity violations” (users will have to flag “lower-severity violations” for review). Zuckerberg also pledged to work closely with the upcoming Trump administration to resist foreign censorship.
The reaction to Zuckerberg’s announcement has been predictably alarmist in traditional media among tech reporters and the sprawling ecosystem of nonprofits dedicated to fighting disinformation and hate speech. Headlines such as “Meta is Not Returning to its Free Speech Origins – It’s Preparing for an Autocratic Future” and “Meta surrenders to the right on speech: ‘I really think this a precursor for genocide,’ a former employee tells Platformer” give you a flavor of the dominant takes. Even international free speech organizations have condemned the move.
There are certainly reasons to be cynical of the timing of and motivation behind Zuckerberg’s announcement, which seems carefully designed to placate and curry favor with the incoming Trump administration. Trump after all threatened Zuckerberg with prison and has long claimed that Big Tech—not least Facebook—has an anti-conservative agenda censoring people on the right.
Yet, if we set aside the political maneuvering and corporate self-preservation behind Zuckerberg’s pivot, the announced policy changes represent a promising shift that could significantly enhance the practical exercise of online free speech. The benefits, however, hinge on Meta resisting co-optation by the very political movement these policies appear designed to appease.
Let’s start with Meta’s move to crowdsource fact-checking. This is not an attack on truth as many fact-checkers—many of whom are paid by Meta—claim. Many fact-checkers do a good job, and undoubtedly approach their calling with the best intentions. But third-party fact-checking has a number of problems. Many official institutions have gotten important calls wrong in recent years, such as the fact-checkers who claimed in the earliest days of Covid that the lab-leak theory was a cranky conspiracy theory, or when Facebook temporarily suppressed the New York Post story about Hunter Biden’s laptop.
The promise of crowdsourced fact-checking, by contrast, is that it avoids the appearance of bias and increases trust by recruiting volunteers to flag misleading posts and provide corrective commentary backed by credible sources. Ideally, an open-source algorithm will ensure transparency by determining which notes are published. The system’s success hinges on incorporating diverse viewpoints and vetting comments through user ratings for clarity, relevance, and reliability. The algorithm then prioritizes notes endorsed by users across political perspectives, minimizing echo chambers and partisan bias.
Some studies suggest that crowdsourced fact-checking can rival expert fact-checking. A University of California, San Diego study of Covid-related “community notes” on X revealed a 96% accuracy rate, with 87% citing high-quality sources. Other studies show that crowdsourcing increases trust in fact-checking. Cornell research, meanwhile, has highlighted how crowdsourced fact-checking in Taiwan, for example, often outperforms traditional fact-checkers at debunking pro-Chinese disinformation.
What’s more, X’s Community Notes feature has held influential figures accountable. In 2022, a White House tweet crediting President Biden for a Social Security increase was flagged with a note clarifying it was due to a 1972 law linking payments to inflation, prompting the tweet’s deletion. Even Elon Musk has been community noted and is currently number 49 on the community notes leaderboard, a website ranking the X users who have been slapped with the most community notes (Musk has notched up 110).
It’s important to note that crowdsourced fact-checking is not a zero-sum game between the crowd and experts. Ordinary users who contribute to community notes rely on experts and rigorous reporting by serious journalists to provide sources that convince other users of the usefulness of the note. In other words, professional fact-checkers and journalists can still contribute to the pursuit of truth, just not as the platform-appointed arbiters of that truth. This is a healthy development, though obviously not a panacea against lies, bullshit and propaganda.
What about Zuckerberg’s justification for these changes—that Meta has “reached a point where [there are] just too many mistakes and too much censorship”?
This too is borne out by the data. Last year my organization, The Future of Free Speech, published a report examining whether social media platforms are overrun with illegal content. If so, we wondered, how are platforms and users moderating that content in response to existing digital regulations?
According to our report, a staggering majority of the content removed from platforms like Facebook and YouTube in France, Germany, and Sweden was legally permissible. We examined deleted comments from 60 of the largest Facebook pages and YouTube channels in these countries, revealing that, depending on the platform and country, between 87.5% and 99.7% of the removed content was legal.
Now, platforms have long banned “lawful but awful” content. But our study showed that over 56% of the deleted comments were merely general expressions of opinion, and not remotely offensive or hateful. This indicates a troubling trend where platforms are sacrificing free speech at the altar of over-cautious moderation, as Zuckerberg admitted. This is also true in the United States. In 2023, social justice activist Shaun King was suspended from Instagram due to “praise for designated entities in violation of our policies” over his posts supporting Palestinians. Such strict content moderation policies also leave little room for irony and satire. Matt Bors, a left-leaning political cartoonist and publisher of The Nib, has frequently faced the ire of Facebook content moderators. The platform removed a cartoon making fun of the Proud Boys for “advocating violence” while another image critiquing President Trump’s pandemic response was taken down for “spreading misinformation” about Covid-19.
In a statement accompanying Zuckerberg’s video, Meta’s new Chief Global Affairs Officer Joel Kaplan—a former Republican operative—asserted that “We want to undo the mission creep that has made our rules too restrictive and too prone to over-enforcement. We’re getting rid of a number of restrictions on topics like immigration, gender identity and gender that are the subject of frequent political discourse and debate.”
One might think it tasteless to single out gender and immigrants—topics that invite conservative ire—but there can be little doubt that Kaplan’s larger point is right. In 2023 we analyzed eight major platforms and revealed a sharp “scope creep” of hate speech policies. Originally targeting things like harmful racism, these policies now cover a broader range of speech, including stereotypes and conspiracy theories, and have more than doubled the number of protected characteristics.
Zuckerberg is also right to point out that speech-restrictive laws in Brazil and Europe have helped fuel this development, with platforms being on the defensive in order to comply with ever more onerous rules. The European Union openly brags about the so-called “Brussels Effect.” Given the trajectory of speech-restrictive laws in democracies around the world, there is a need to push back against such regulation. If Meta is sincere about reversing these trends while simultaneously giving users more agency to control what type of content they want to view, this is good news for free speech. And if these policies are implemented in good faith, it will also mean giving users freer rein to criticize Trump and a Republican-dominated Congress that promises to implement policies unpopular with many liberals and independents.
The real danger with Meta’s pivot toward Trump isn’t an excess of free speech—it’s the potential risk that Zuckerberg and other tech moguls will be co-opted by the upcoming Trump administration. The emphasis on free speech could become a smokescreen for an arranged marriage between Silicon Valley and the White House. Such a scenario could lead to content policies favoring MAGA ideology while suppressing dissent, or even collusion with the administration to silence critics through secret backchannels, a practice known as “jawboning.” If this dystopia becomes reality, it will be fair to warn against an “autocratic future,” as many are currently doing.
But until then, while keeping a sharp eye on how it is implemented, those who care about free speech should welcome Meta’s U-turn.
Jacob Mchangama is the Executive Director of The Future of Free Speech and a research professor at Vanderbilt University. He is also a Senior Fellow at The Foundation for Individual Rights and Expression (FIRE) and the author of Free Speech: A History From Socrates to Social Media.
Follow Persuasion on Twitter, LinkedIn, and YouTube to keep up with our latest articles, podcasts, and events, as well as updates from excellent writers across our network.
And, to receive pieces like this in your inbox and support our work, subscribe below:
Another platform that needs reform is Amazon. Most people don't know that Amazon reviews may be censored without valid reasons or any recourse at all, especially book reviews. And not just particular reviews, but all reviews, past and future, for unspecified reasons, which could easily be based on the Amazon censor's ideology, politics, biases ,or ignorance.
Hell no it is not up to Zuck to "keep Trump at arm's length". That is the illiberal Constitutionally-incorrect issue that Zuck is fixing. It is not up to platforms to moderate content other than simple common decency as defined by current FCC regulations and what children have access to. The consumers of content are the only control source for what they ingest and believe.