The End of the Open Internet

At this year’s Munich Security Conference, European Commission President Ursula von der Leyen stressed Europe’s “long tradition in freedom of speech.” Then she drew a redline. “We are very clear with digital sovereignty . . . that what is forbidden offline is forbidden online,” she added.

Von der Leyen’s maxim would have sounded foreign to the continent’s leaders just over a decade ago. Inspired by the social media–led movements of the Arab Spring, liberal democracies treated Internet freedom as a geopolitical principle to be evangelized rather than a problem to be regulated. Since then, faith in the liberalizing potential of open access to the Internet has given way to a more technocratic focus on digital sovereignty, the idea that states must control their own data and infrastructure, as the organizing principle of European digital policy. The pivot has come as a response to the increasing dominance of American tech platforms, whose engagement-driven models have driven fears that they might be weaponized by hostile states and groups to spread propaganda and undermine democratic institutions. Since U.S. President Donald Trump’s reelection in 2024—and as Silicon Valley’s leaders have cozied up to an administration openly antagonistic toward Europe’s political establishment and supportive of the populists challenging it—European policymakers have increasingly felt the need to assert control over what Spanish Prime Minister Pedro Sánchez has referred to as a “digital Wild West.”

Europe has already committed to scaling its own technological capacity and reducing dependence on the U.S. tech stack through measures such as the European Chips Act, which aims to boost semiconductor production, and projects such as Gaia-X, a cloud infrastructure initiative designed to facilitate secure data transfers. But the continent’s adoption of digital sovereignty now extends well beyond industrial policy and critical infrastructure. Digital sovereignty has come to encompass the governance of speech, as governments push social media platforms to police online expression vaguely classified as disinformation, foreign manipulation, hate speech, or child exploitation.

In the name of safeguarding democracy, open societies are importing the policies of authoritarian regimes they rightly identify as enemies of free expression. The European Union and its member states have embraced top-down content controls. Meanwhile, the Trump administration has expanded its own surveillance, ideological screening, and pressure on tech platforms. It is not too late to reclaim the open Internet’s promise of uninhibited access to information across borders. But this requires making a distinction between holding tech platforms accountable and imposing state censorship—and decisively rejecting the latter while encouraging platform design that ensures the former.

A DIGITAL DREAM DEFERRED

In July 2012, the UN Human Rights Council (UNHRC) unanimously adopted a landmark resolution affirming a simple proposition: “The same rights that people have offline must also be protected online, in particular freedom of expression, which is applicable regardless of frontiers and through any media of one's choice.” Writing in The New York Times, Carl Bildt, Sweden’s foreign minister, hailed the vote as a “breakthrough of fundamental importance,” insisting that “the free flow of information on the Internet is a global call and not something pushed only by a few Western states.” For a brief moment, it seemed as if the world’s democracies had accepted that the open Internet would be the nervous system of modern liberty.

But even at this crowning moment of techno-optimism, that idea was already facing challenges. After pro-democracy activists in Russia used social media to coordinate mass protests against fraudulent legislative elections in 2011 and the presidential election in 2012, the Kremlin treated online dissent as a national security threat. The Federal Security Service, or FSB, accused Russian bloggers and American social networks of disseminating Western calls for “overthrowing the established political regime.” The same month that the UNHRC resolution was approved, the Duma passed what was nominally a child protection law that in practice created a sweeping national blacklist authorizing the media regulator Roskomnadzor to block without a court order any sites deemed to be promoting drug use, suicide, and ultimately anything classified as “extremism.” The law laid the foundation for Russia’s “Red Web,” the system of state-directed censorship and surveillance through which the Kremlin has progressively isolated Russians from independent information, silenced political opposition online, and entrenched Russian President Vladimir Putin’s grip on power.

China’s efforts to control digital spaces have been even more ambitious. A year after Xi Jinping rose to the top of the Chinese Communist Party in 2012, the Central Committee circulated what became known as “Document No. 9,” a communiqué warning that Western constitutional democracy, “universal values,” civil society, and press freedom could “penetrate” China through the Internet. The remedy was twofold: ideological enforcement through “unification of thought” and what the document called the “purification” of online public opinion. In practice, it meant significantly extending China’s “Great Firewall”—systematically blocking Facebook, Google, Twitter, and most Western news outlets—while adding new layers of real-name registration requirements, instituting algorithmic keyword filtering on domestic platforms such as WeChat and Weibo, and surging the arrests of bloggers and activists for online speech. The CCP has since extended this governance structure to China’s AI companies, which are under strict requirements to censor all outputs that conflict with “core socialist values.”

Europe and the United States have undergone shifts of their own since the 2012 resolution. In Europe, fears that Russian disinformation swayed the 2016 U.S. election, an “infodemic” of false and misleading public health claims during the COVID-19 pandemic, and both Trump administrations’ support for European populists and hostility to EU Internet regulation have propelled policymakers’ embrace of digital sovereignty. And in the United States, the second Trump administration has expanded executive power and pursued punitive action against critics, while decrying what it perceives as Europe’s censoriousness.

PANIC AT THE DISINFO

The 2012 UN resolution stressed the importance of protecting the same human rights online that apply offline, especially freedom of expression. But now, European leaders are racing to add restrictions to online speech. In February, German Chancellor Friedrich Merz advocated the end of online anonymity. “I want to see real names on the Internet. I want to know who is speaking,” he said at a Christian Democratic Union event in February. Merz’s proposal—deeply at odds with the Enlightenment tradition embodied by Spinoza, Voltaire, and Locke, all of whom relied on anonymous publication to defy censorship—would enable even greater speech restrictions than Germany already imposes. Investigations of Germans for insults against political figures rose from 2,598 in 2023 to 4,792 in 2025, an increase of some 85 percent. In early June a German Facebook user was fined for calling Merz a “lying Fritz,” and other Germans have been convicted for similar criticism of senior cabinet ministers.

Elsewhere on the continent, French President Emmanuel Macron has dismissed as “pure bullshit” the idea that algorithmic distribution of content should qualify as free speech and stressed the need for new laws to protect democracy against foreign propaganda. Macron has also called for urgent measures to block “false information” or “information harmful to a person’s dignity” on social media, which would represent a dramatic expansion of a 2018 French law that empowers judges to order the removal of false online information during election periods.

Other European countries, including Austria, Denmark, and Spain, have also expanded their national hate speech laws, and the EU is proposing to make hate speech a crime across all member states. The expansion of broad and ambiguous categories of illegal content under both national and European law has significant ramifications, since the EU’s 2022 Digital Services Act obliges social media platforms to remove such content. In other words, the more speech that is prohibited by member states and the EU, the more obligation placed on tech companies to police the Internet—and the more likely those companies are to restrict, rather than strengthen, the voices of users to avoid punitive fines and lengthy investigations by EU institutions.

The target of these initiatives would appear to be tech billionaires, whose platforms can fuel polarization, supercharge content that no responsible editor would publish, and keep children scrolling on screens. The power of private platforms is indeed a legitimate worry for Europe. A handful of foreign-owned companies do help shape what speech is amplified and suppressed. But the way states are responding creates a more durable threat than any individual owner of a tech company. Once built, censorship infrastructure is difficult to tear down. Most likely, it will be inherited by every future government—including those controlled by the right-populist parties currently surging in polls across the continent.

And those who will suffer are not the billionaires in Silicon Valley, but the millions of ordinary users of platforms in European democracies who access, share, and discuss information and ideas about topics that make their elected representatives uncomfortable.

The fear of foreign interference through social networks also has led policymakers to underestimate Europe’s democratic resilience and capacity for critical thinking. European governments tend to exaggerate the reach and effect of online disinformation on elections and its ability to durably change deeply held views. A growing body of empirical research has found that exposure to online misinformation is concentrated among a small minority of already-committed partisans and that its direct persuasive effects on voting behavior are modest. European support for populist parties, therefore, likely has less to do with MAGA messaging or Russian disinformation on X than with organic, genuine concerns over immigration and the economy that have mobilized anti-elitist voters against institutionalist incumbents for years, long before Trump became president and Twitter became X.

Diversity of opinion and criticism of political decisions, whether domestic or foreign in origin, lie at the very heart of free speech. Labeling such criticism as an attack on democracy implies that democratically elected leaders’ policies and priorities should be insulated from scrutiny. Although Merz and others have described their efforts as targeting the “enemies of an open and liberal society,” their proposed prohibitions bear a striking resemblance to policies of illiberal regimes who use control of the Internet to retain power and throttle dissent. It is no accident that real-name identification rules are enforced in China, Iran, Russia, and Vietnam.

TRANSATLANTIC DRIFT

The United States offers no reliable counterweight to Europe’s alarming drift. The Trump administration has vociferously criticized the EU’s digital speech governance, but the United States’ commitment to Internet freedom under Trump is at best inconsistent. U.S. agencies have leaned on tech and social media platforms to unmask anonymous speakers and remove user reporting on the activities of immigration enforcement officers. And they have expanded the vetting of visitors and students far beyond standard security screening to include reviews of social media accounts for their ideological content. When the AI company Anthropic refused to let the Pentagon use its products for the mass surveillance of U.S. citizens or the production of autonomous weapons, the administration designated the company a “supply chain risk” and ordered all federal agencies to cut ties. Litigation is ongoing, but in March a federal judge issued a preliminary injunction in Anthropic’s favor, calling the designation a “classic First Amendment retaliation” against protected speech.

In foreign policy, the contradiction between the United States’ critique of Europe and its own actions is even starker. In February, Reuters reported that the State Department was building a website giving Europeans access to content blocked under EU and national laws. Yet it offered no comparable effort for citizens of China, Iran, or Russia facing vastly harsher censorship without any of the democratic recourse available to Europeans. In fact, Washington has quietly scaled back Internet freedom initiatives, including the Open Technology Fund, originally designed to help people in authoritarian states circumvent online censorship.

To be sure, Washington's and Brussels' approaches to online speech are hardly symmetrical. The Trump administration’s actions, ad hoc, executive-driven, and retaliatory as they might be, are nonetheless constrained by First Amendment protections that Europeans largely lack; Europe’s speech laws, although often vague in scope and opaque in enforcement, are the result of deliberation by elected institutions. Still, both are united by their divergence from the ambitious commitments of the 2010s and by their refusal to offer a truly democratic alternative to authoritarian digital sovereignty.

POWER TO THE PEOPLE

Looking back, it is difficult to avoid the conclusion that in the clash between the Western-led open Internet agenda and the Chinese-Russian internet sovereignty model, the latter is winning. The digital space of Western democracies is vastly more open than what is allowed past China’s Great Firewall or Russia’s Red Web. But in the United States and Europe today, the 2012 UNHRC resolution affirming online rights reads like a relic from a bygone era. Many of its principles have been turned on their head, and Bildt’s techno-optimistic blueprint for a global and open Internet has been quietly shelved.

The real promise of the open Internet was never that platforms would be impartial gatekeepers or that providing everyone with free and immediate speech at scale would be frictionless. It was that ordinary citizens would have a more direct voice in public affairs and a better chance of holding their governments accountable. In reality, this ideal was always going to be messy, not messianic.

For all the concern over X’s antidemocratic potential under the ownership of the Trump-aligned tech titan Elon Musk, the platform provided a powerful example of this promise in practice. After bystander video of the killing of Alex Pretti by U.S. immigration agents in Minneapolis spread on X, users turned to the site’s Community Notes to quickly append corrections to false and misleading claims by senior government officials, including White House Deputy Chief of Staff Stephen Miller, then Homeland Security Secretary Kristi Noem, and FBI Director Kash Patel, as well as the official account of the Department of Homeland Security. As a result, the Trump administration had no choice but to walk back its initial attempt to flood the zone.

In Europe, the Hungarian opposition’s recent victory over Prime Minister Viktor Orban should also make Brussels think twice about digital sovereignty. After more than a decade of consolidated control over traditional media, Hungary’s remaining independent journalists and embattled opposition relied heavily on the Internet and social media to hold the corrupt and illiberal government to account and ultimately to mobilize the votes that ended Orban’s 16-year rule.

Such organic, bottom-up responses are precisely why authoritarian regimes fear connectivity. When Iran crushed protests in January, the regime imposed digital darkness, shutting down the Internet across the country to prevent the world from bearing witness. Iran’s UN ambassador blamed the protests on externally orchestrated terrorism and asserted the regime’s “sovereign right” to stop “unauthorized transmissions.” The blackouts were the ultimate display of government control over the flow of information—one that should serve as a warning to Western policymakers about where the idea of digital sovereignty logically leads.

For European democracies, there is an alternative to top-down digital sovereignty. A more decentralized, interoperable and pro-social online ecosystem designed to bridge political and geographic divides rather than amplify them does not have to choose between democratic accountability and online safety. Governments concerned with both can protect user privacy and encourage crowd-sourced fact checking and transparent platform design that empowers users rather than the platforms themselves. They can enforce existing laws against fraud, incitement, and child exploitation without building a new censorship infrastructure. They can, in short, take the 2012 resolution at its word by protecting human rights online instead of extending government control into the digital public square.

Loading...