This essay is part of a series from Carnegie’s digital democracy network, a diverse group of thinkers and activists engaged in work on technology and politics. The series is produced by Carnegie’s Democracy, Conflict, and Governance Program. The full set of essays is scheduled for publication in summer 2026.

On March 10, 2026, Iran’s government spokesperson Fatemeh Mohajerani made a notable public admission in the middle of a war, at a moment when its citizens had already spent one-third of 2026 in a near-total digital darkness thanks to shutdowns imposed by the regime. She said, “For those who can carry our voice further, opportunities will be provided.” This is a reference to the regime’s privileged  system that grants unfiltered connectivity to select individuals. Five days later, Foreign Minister Abbas Araghchi appeared on CBS News via Zoom while millions of Iranians remained offline. He was asked why he had internet access. His answer: “because I’m the voice of Iranians.”

Together, these statements did something new: Rather than hiding behind security justifications, regime officials openly defended connectivity as a political instrument. The internet is reserved for those who carry the state’s voice, and it is withheld from everyone else. Internet access during a dangerous war was not being restored, rather, it was being allocated exclusively to people who would amplify the regime’s voice outward.

These statements point to something that has received little attention from those looking at authoritarian information controls: Connectivity has become content policy. The pipeline and what flows through it now constitute the same political act. Understanding Iran’s strategy requires looking past the regime’s censorship and asking what it is building and why.

A Legitimacy Problem, Not a Censorship Problem

The standard framework for understanding Iran’s information controls treats the regime’s tactics primarily as censorship: blocking, filtering, surveilling, flooding, and shutting down connectivity. Iran has spent two decades investing in building the infrastructure for information suppression, an effort that accelerated dramatically since the Woman, Life, Freedom uprising of 2022. But censorship, or the subtractive idea that the regime is taking something away, only partially explains what is happening in Iran. The more important question is: What is the regime constructing in place of an open internet, and why does it need to build it now?

The regime’s survival strategy depends on coercion and fear. Authorities have met the uprisings of 2019, 2022, and 2026 with militarized crackdowns of escalating brutality. Spreading fear and manipulating information are not mutually exclusive: they are layers of the same system, each enabling the other. The surveillance identifies who to target. The shutdown covers up the killing. The propaganda and disinformation fills the silence the shutdown creates. Conditional internet access grants connectivity only to those deemed loyal by the regime and capable of serving its interests. Together, these strategies form a unified architecture of repression, assembled in response to a specific and deepening problem: declining legitimacy.

Each successive uprising in Iran has voiced a diversity of demands while intensifying consensus against the regime. The 2019 protests drew in the urban and rural poor, ground down by fuel price rises and economic mismanagement. The Woman, Life, Freedom movement of 2022 brought in the middle class, students, and ethnic minorities: a cross-class coalition not seen since 1979 that demanded an end to an oppressive system. The 2026 uprising went further still, with millions taking to the streets, including the merchant class, frustrated by the country’s economic collapse. The latest protest so alarmed Iran’s security state that it reacted with unprecedented brutality, arresting more than 40,000 and killing thousands, with the estimated death toll up to 20,000 or even 30,000. Each cycle of protests has drawn in broader portions of society, leaving the regime with a shrinking base of genuine support.

The erosion of support for the Islamic Republic has now collided with an external crisis that the regime has every incentive to exploit. On February 28, 2026, the United States and Israel launched joint strikes against Iran with shifting objectives: degrade Iran’s missile capabilities, eliminate its nuclear program, or even bring about regime change. The effect has been to push the Islamic Republic into an existential confrontation unlike any since the Iran-Iraq War. For a regime already facing the gravest legitimacy crisis in its history, these attacks may also offer authorities an ideological lifeline.

The Islamic Republic’s founding claim, written into Article 154 of its constitution, commits it to support “the just struggles of the mustad’afun [oppressed] against the mustakbirun [tyrants] in every corner of the globe.” That is, Iran stands as the voice of the oppressed against imperial aggression. That claim has rarely had more raw material to work with. With civilian casualties mounting from foreign strikes, the regime now has authentic evidence to offer international audiences, including diaspora communities, Global South solidarity networks, and the anti-war left, about Western aggression. The information architecture it has spent years building is now oriented outward when the ideological stakes are highest.

That the regime is now the target of very real foreign aggression, not just the biggest oppressor of its own people, helps it to assume the victimhood it has always claimed. The policy Mohajerani was articulating is directed at two audiences. Externally, Iran is attempting to convince the world about the regime’s persecution. Internally, it is saturating the information sphere with propaganda, exploiting the vacuum left by silenced opposition voices.

When Iran’s spokesperson says connectivity will go to people who can “carry our voice further,” she is not describing a reward for loyalty. She is describing a transaction.

This is the problem that Iran’s information control architecture exists to solve: not just to silence dissent, but to construct the appearance of popular support, which the regime no longer commands in reality. Each uprising has converted more of the population from passive acquiescence into active opposition. When Mohajerani says connectivity will go to people who can “carry our voice further,” she is not describing a reward for loyalty. She is describing a transaction: connectivity in exchange for amplification.

That is not a censorship problem. It is a legitimacy problem that has been building for forty-seven years, and the information architecture is the answer.

Iran’s Information Architecture, Built Layer by Layer

The foundation of the regime’s information architecture is surveillance. Before the regime can shut down selectively, flood strategically, or allocate internet access based on loyalty, it needs to catalog its population—who carries influence, which constituencies do they represent, and whether they can validate the state’s legitimacy. Iran’s surveillance infrastructure, built with significant assistance from Chinese technology companies including Huawei and ZTE, enables deep packet inspection and the extraction of user data from platforms. But as Afsaneh Rigot’s research documents, surveillance on the ground is blunter. Device inspections are routine. They are used at checkpoints and after arrests or to wipe documentation of protests, identify anti-regime sentiment, and criminalize mobilization before it can spread. The result is the creation of an intelligence architecture that supplies the regime with the power to know precisely where to apply pressure and where to extend trust.

The shutdown is the government’s bluntest instrument, deployed when opposition to its power is mobilized. Iran has been refining this capability since 2009, when it imposed its first significant internet restrictions during the Green Movement protests. Ten years later, by November 2019, the infrastructure had matured enough to execute a near-total nationwide shutdown of internet traffic and a severing of users from global networks in response to nationwide protests. A Reuters investigation put the death toll at roughly 1,500, a figure that remains contested because the shutdown made it impossible to know the true measure of the regime’s brutality. This pattern has repeated with every subsequent uprising. In 2026, the regime enacted its most brutal crackdown yet, this time under the cover of a total communications blackout that severed the internet alongside phone networks and landlines. The blackout meant one thing to those familiar with the regime’s methods: carnage in the darkness.

But the shutdown does not simply erase the information environment. Instead, it creates a vacuum which the regime has learned to fill.

The content flooding layer reached new prominence during the Woman, Life, Freedom uprising of 2022 and reached industrial scale with the advent of generative AI during the twelve-day Iran-Israel war of June 2025. This author with Sam Gregory published research documenting how AI-generated content has taken disinformation to an unprecedented level. Synthetic videos of pro-Israel protests in Tehran falsely suggested popular dissent against the regime, while fabricated images of Israeli F-35 planes shot down over Iran were broadcast on state television portraying the state as victorious. The information space was flooded in every direction at once, making it difficult to discern what was actually happening.

But flooding is indiscriminate. A more precise tool, is what Iranians call “cyberis,” regime-operated accounts that infiltrate the opposition’s channels by building long-standing followings as trusted anonymous opposition voices which are then deployed to redirect or corrupt information at critical moments. In January 2023, protesters were already mobilizing to prevent two executions when a well-established opposition account spread the claim that Iran’s notorious “hanging judge” had been assassinated. The story consumed the attention of Iranians and international observers alike. The executions were carried out quietly while everyone was looking elsewhere. More recently, user accounts impersonating opposition voices created manipulated protest content and used it to cast doubt on authentic footage, seeding claims that footage of escalating demonstrations in January were fake. The goal in both cases was not to silence the opposition but to speak in its voice and use that voice against it. In doing so the regime has produced a consequence it may not have fully planned but has certainly exploited: the liar’s dividend.

The liar’s dividend, in which bad actors strategically claim that stories are false thereby undercutting public trust in legitimate information, cuts both ways. The regime spent years contaminating the information environment so thoroughly that when it most needs its victimhood claims believed—that foreign strikes are responsible for mounting civilian casualties—they are no longer viewed as credible. The white SIM infrastructure is the regime’s attempt to solve that problem—authorities offer special, unfiltered communications access to designated individuals who can bypass the national internet filtering system. This system has existed in various forms since at least 2013. It was formally instituted by the Supreme Council of Cyberspace in July 2025, and ballooned into a public scandal in November 2025 when X’s location feature exposed pro-government figures posting from Iran without VPNs during an internet blackout. The controversy revealed something more troubling than mere privilege. It showed that figures long treated as opposition voices were using their white SIM access to manipulate online narratives in favor of regime interests while publicly claiming persecution.

While Mohajerani dismissed the concept of a tiered internet as recently as July 2025, saying the government’s approach was “free internet,” her March 10 reversal finally revealed the state’s true intent: allocate online access only to those people who will reinforce the regime’s interests.

What This Means for the Response

The international community’s responses to Iran’s information crisis share a common premise: that the information environment is something to be defended or restored. Internet freedom organizations seek to reconnect people who have been cut off. Digital security tools and guardrails work to protect against surveillance and device seizure. Verification bodies and detection tools identify synthetic and fake content. Platform accountability frameworks push for labelling and moderation. Each is legitimate within its own frame, but they share a faulty premise: that infrastructure and content are separate problems. Mohajerani’s statement collapses that distinction. The decision about who gets connectivity is now also the decision about what gets said. Content policy has moved upstream, into the pipe itself. The fight is no longer only over what people can say. It is over whether they are allowed to speak at all, and what is constructed in their silence.

Traditional terrestrial internet gives states a chokepoint—national sovereignty over infrastructure becomes sovereignty over voice. But there are alternatives. A promising approach is satellite internet, not as a circumvention tool, but as a challenge to the premise of state information control entirely. The Direct2Cell coalition that is organized around promoting satellite connectivity is based on this exact logic: bypass the chokepoints the regime has spent two decades building and render the conditional access apparatus at least partially unenforceable. But free connectivity is not sufficient if it replicates the same concentrations of power in different hands. As Rigot documented during the January blackout, Starlink is toggleable at a single company’s discretion, available only through smuggled dishes, and carries serious security risks for anyone whose device is later searched. The answer to state control of connectivity is not private control. It is democratic international governance, satellite internet treated as a global commons, accountable to the people whose lives depend on it. The Iranian Women’s Coalition for Internet Freedom (of which this author is a co-initiator) stated in its March 2026 manifesto that satellite internet governance is internet governance, and the institutions overseeing it must be accountable to people, not only to states and private actors.

Iranians are dying under foreign strikes. The regime did not manufacture those deaths. But it has spent decades building the infrastructure to control how they are seen.

Iranians are dying under foreign strikes. The regime did not manufacture those deaths. But it has spent decades building the infrastructure to control how they are seen, and Mohajerani’s announcement is the moment that the regime openly declared what its infrastructure was designed to do. The people with white SIMs are not being given access to communicate. They are being given access to broadcast.

Araghchi’s CBS appearance was that operation made visible. A foreign minister conducted international influence work on American television via Zoom, while his own citizens could not securely call their families. This is not an anomaly. It is the architecture functioning exactly as intended: connectivity allocated not as a public utility but as a transmission belt for the state’s voice outward, toward the diaspora, the Global South, and the international community. The regime has concluded it no longer needs to win the information war inside Iran. It only needs to win it outside.

What travels through the pipeline, who carries it, and who is silenced in the same moment are not separate questions about content and connectivity. Interventions that treat them as separate will always arrive too late. This is an infrastructure problem. Until the international community names it as such, the architecture will be complete before anyone has agreed on what to call it, let alone how to stop it.