Meta’s Move to Limit Fact-Checking Endangers Women—and Democracy

“We’re going to catch less bad stuff.” That’s how Mark Zuckerberg described his recent decision to end Meta’s fact-checking program, conceding that more harmful posts are going to make their way onto his platforms. What he failed to acknowledge or account for, however, is the repercussions this will have for women’s well-being, safety, and ability to participate fully and freely in public life. The tech platform’s move to curtail fact-checking and remove restrictions on specific topics—Zuckerberg singled out “gender and immigrants”—signals a profound step backward that will disproportionately harm women, and LGBTQ+ and minority communities, and amplify the spread of hate and misogyny. Research from the Georgetown Institute for Women, Peace and Security (GIWPS) further finds that widespread digital violence against women fundamentally erodes American democracy. 

GIWPS’s research draws on an extensive body of work demonstrating that women’s experience of the digital world is marked by pervasive violence. Even with some content moderation, women face disproportionate levels of violence in digital spaces—the Economist Intelligence Unit estimates that 85 percent of women globally experienced or witnessed gendered violence online. This violence takes many forms, ranging from sexual and misogynistic threats on social media and the use of artificial intelligence to create non-consensual sexually explicit images, to coordinated disinformation campaigns relying on gender stereotypes, designed to discredit women and undermine democracy through the spread of false information. These threats should be taken seriously; the violence women face in digital spaces impacts their well-being, mental health, and safety. As an additional layer, these threats erode free and equal participation in public and political spaces, leading to a degradation of our democracy. Digital violence replicates and reinforces inequalities and exclusions we see in the offline world. 

This issue flourishes at home and abroad. Last year, former Vice President Kamala Harris became the most targeted American politician on the internet, according to a Wilson Center report, as she faced more toxic discourse than Joe Biden or other male Democratic candidates before her. U.S. Rep. Alexandria Ocasio-Cortez (D-N.Y.), who has been a persistent target of online harassment since she ran for Congress in 2017, was one among 25 lawmakers found to be the subject of non-consensual sexual imagery;  while scrolling X (formerly Twitter) one day, she discovered a digitally altered image of someone forcing her mouth on their genitals. This image of Ocasio-Cortez is just the tip of the iceberg. Research by the American Sunlight Project found 35,000 mentions of 26 women in Congress on deepfake sites. Ocasio-Cortez led a bipartisan effort to introduce the Disrupt Explicit Forged Images and Non-Consensual Edits Act of 2024 (DEFIANCE Act) to protect victims of non-consensual images, which passed the Senate in July 2024 but was never passed in the House. Women of color and LGBTQ+ people experience the most and the worst of online violence, though more research is sorely needed. Though these types of attacks are more frequently documented against Democratic women, Republicans in public life do not survive unscathed. Former UN Ambassador Nikki Haley faced a barrage of offline and online sexist attacks, with X users saying she will be “remembered as a bootlicking c*nt” and is an “ugly b*tch who loves money [...] that f*cking ugly wh*re.” Women in the military, on the campaign trail or serving in local and statewide office, and election workers also face disproportionate threats, like “you should be shot for treason” and “Your daughter is beautiful... I’d be a shame if something happened to her[.]”

Around the world, women leaders face an uglier, more violent internet than their male counterparts: During the Indian general election in 2019, one in seven tweets about female politicians was abusive; 42 percent of women politicians in the African Union have received online rape or death threats, and across Southeast Asia, 50 percent of women politicians faced online abuse. Kavita Krishnan, a politician and activist in India, says she regularly receives messages telling her, in her own words: “I’m not worth raping, what kind of torture and rape I should be subjected to, telling me what kind of men I should be sleeping with.” Female journalists and activists are also subject to gendered violence intended to muzzle and discourage them from holding positions of power. While the immediate impact is harmful, the repercussions aren’t limited to the digital world. Online attacks often escalate to offline violence—the murder of British parliamentarian Jo Cox, for example, was famously preceded by a flood of online abuse.

These instances of abuse happened even before the removal of Meta’s bare-bones regulations, and women’s groups have come together to push for more strenuous safeguards. In one instance, these groups made calls to action in 2020 after a manipulated video (not the first) began circulating on social media that showed then-House Speaker Nancy Pelosi appearing to be drunk. Meta declined to remove the altered video and instead attached a “partly false” label to the clip, though experts on the spread of misinformation found that such labels alone are not sufficient to stem the spread of false information. Meta’s latest decision to further loosen moderation rules removes critical safeguards against the spread of harmful content, particularly for women and members of marginalized communities. Zuckerberg referred to corporate culture at Meta as becoming “neutered” and said the new policies would welcome back masculinity and aggression. He described the changes as “simplifying content policies” on the topics of immigration and gender, but in reality, the guidelines are opening the floodgates of hateful and degrading language. Examples of comments explicitly allowed under the new policy include those likening women to property, calling transgender people mentally ill, saying that “gays are freaks,” and that immigrants are “filthy pieces of sh**.” These recent rollbacks remove the already extremely limited guardrails protecting women and LGBTQ+ people from abuse and violence in digital spaces. 

Though the pursuit of a just and equitable digital ecosystem for all is a worthwhile endeavor in and of itself, rampant online violence against women has a “chilling effect” detrimental to a healthy democracy. In the context of digital spaces and social media platforms, a chilling effect occurs when people self-censor or avoid the platform altogether due to fear of repercussions and violence. When women choose to opt out of these conversations, it worsens the quality of public discourse in ways that are only compounded by the role that algorithms play in boosting and repressing different political narratives, and it limits the issues that are discussed in online fora. In the United Kingdom, 73 percent of female members of parliament (MPs)—compared to 51 percent of male MPs—decide not to speak about certain issues online, and in Argentina, activists described online violence during debates about abortion as a “muzzle.” The issues women choose not to speak about are frequently those, such as reproductive rights, that relate directly to the advancement of gender equality and human rights more broadly. Their silence prevents these issues from pushing forward. 

Other women choose to leave office or not run altogether. A UN Women report found that across India, Pakistan, and Nepal, 60 percent of women did not participate in politics due to fear of violence. In Brazil, two congresswomen publicly announced that they would not run in the 2022 election after being targeted by digital violence. Women don’t even have to be the direct victims of online violence to be discouraged from participating —PLAN International found that one in five women aged 15-24 have stopped engaging in politics and civic spaces due to fear of being targeted online. In a world where digital spaces are so tightly intertwined with both people’s daily lives and politics in general, the exclusion of women from online spaces constitutes a challenge to truly equal participation, the bare minimum requirement for a democracy. The inclusion of 50 percent of a country’s population is a definitional requirement for a democratic country, and GIWPS research shows that the status of women and gender equality is strongly and significantly associated with a healthy democracy. Our findings suggest that in countries where women are doing well, there are more likely to be fair elections, autonomous civil societies, and independent and accountable government institutions.

Indeed, foreign actors seem to be aware of this connection. “It’s a strange group of malign actors, who have come together to use gender equality as the tip of the spear to undermine democracy,” said Geeta Rao Gupta, who served as ambassador-at-large for global women’s issues during the Biden administration. While some forms of digital violence are carried out by individuals motivated by nothing more than misogyny, many attacks are part of coordinated, well-funded campaigns by malign actors. Russia is notorious for its gendered disinformation tactics, most recently carrying out an extensive campaign in Moldova that exploited beliefs about stereotypical gender roles to undermine the credibility of President Maia Sandu. Former Moldovan President Igor Dodon, who has been widely criticized for his pro-Kremlin stance, insinuated as part of his campaign against Sandu that “If you didn’t manage to build a family, how can we expect you to build a country?” These attacks were not just personal—they were designed to undermine and discredit Sandu’s leadership, with the ultimate intention of derailing attempts at a referendum on EU membership. Kristina Wilfore, co-founder of #ShePersisted, a global platform for addressing digital threats to women leaders, explained: “This is no ordinary election. It’s an all-out assault on Moldova’s sovereignty, and at the heart of the battle lies one simple truth: Whoever controls the narrative determines Moldova’s future.” 

The Chinese Communist Party (CCP) is also known to use digital violence as a means of consolidating power and repressing dissent, retaliating against diaspora women who criticize the CCP, using accounts with fake identities to inundate them with vicious gendered attacks, and collaborating with misogynistic ultranationalist users to attack feminist activists, accusing them of being anti-China, “sluts,” “race traitors,” and foreign agents. The Taliban also uses social media to harass women, accusing them of promiscuity and alleging the women are agents of the West. Social media is also weaponized to exacerbate conflict and even genocide, for example, in Myanmar, where attacks against the Rohingya people were promoted, and military leaders actively used the platform to inflame public sentiment; one posted a false story about the rape of a Buddhist woman by a Muslim man. Similarly, ethnic tensions against the Tigray community were stoked in Ethiopia, with Amnesty International finding that Facebook “supercharged the spread of harmful rhetoric.” In both places, sexual violence was legitimated. In light of these trends, Meta’s decision to end fact-checking is all the more alarming; while each of these examples was not fully contained by fact-checking processes, research has consistently shown that fact-checking is an important and effective mechanism for curbing the spread of misinformation. Recognizing that violence persists despite fact-checking, the response should be to refine and strengthen these tools—not to dismantle the limited protections that exist. 

It is no coincidence that Zuckerberg’s decision came just as Donald Trump—who has trafficked in the spread of misinformation with repeated attacks aimed at women and LGBTQ+ people—returned to the White House. When asked if Meta’s policy changes came in response to threats he had made against the company and Zuckerberg,  criticizing them for “[steering against him]” in the 2020 election, Trump said, “Probably.” Harvard Kennedy School’s Erica Chenoweth and Zoe Marks argue that rising authoritarianism and rollbacks on women’s rights are “mutually reinforcing ills,” given that historically the fight for women’s and civil rights and a healthy democracy have gone hand in hand. Women’s participation in social movements increases the likelihood that they succeed and is more likely to lead to a more egalitarian democracy. Fully free, politically active women, they argue, are a threat to authoritarian and authoritarian-leaning leaders.

Meta’s rollback is not a neutral policy shift—it is an abdication of responsibility that puts women, minority communities, and democracy at risk. Research consistently demonstrates that women’s leadership contributes to more equitable policymaking, reduces corruption, and strengthens democratic institutions. Women, in particular those from marginalized communities, seeking political office in 2024 faced intensified risks. Online threats drive women to disengage from public discourse, self-censor, or withdraw from political and professional opportunities altogether. 

Addressing these challenges is critical. What can be done? One promising model is Australia’s eSafety commissioner, which leverages a proactive approach—mandating swift takedowns of abusive material and providing practical support for victims—which are lessons for other nations grappling with the dark side of digital spaces. Equally, the European Union’s Digital Services Act (DSA) has set a high standard for tackling disinformation. With a newly strengthened Code of Practice, the DSA compels major platforms to work to prevent and address the spread of hate speech and work with researchers and fact-checkers to combat the spread of disinformation. 

It is not surprising that Zuckerberg is attempting to ensure that comparable measures are not adopted here in the United States or allowed to impact his overseas operations. As a result, there will be even greater controversy surrounding the statutory protection that safeguards digital platforms against liability for content posted by third parties. This provision enables platforms to abandon any efforts to moderate content. Greater attention should be paid by both activists and lawmakers to whether Section 230 of the Communications Decency Act strikes an appropriate balance between free expression and responsibility for harmful content.

In a context where it seems that the government is complicit if not compliant in the deregulation of digital spaces, turning to and supporting women-led civil society organizations can be vital in advancing safety and equality in the digital realm. Organizations such as Chayn, which provides online services for survivors of abuse, #ShePersisted, which conducts rigorous research to help women leaders build digital resilience, and Women’s Rights Online, which brings together organizations working on gender justice and digital rights to coordinate efforts, fill critical gaps left by mainstream initiatives. As we advocate for safer digital spaces, it is crucial to engage grassroots movements as essential partners. 

The decision to catch “less bad stuff” may not concern Mark Zuckerberg—and may be welcomed by Trump, which is the point. But women, members of marginalized and vulnerable communities, and democracy will pay the price, and the worst thing we, as civil society, can do in the face of such blatant attacks on democracy is nothing.