When Content Standards Cannot Be Unified

Global platforms face a structural contradiction: information crosses borders, but the rules governing it do not.

For Meta, TikTok, and WeChat, the problem is no longer simply which content should come down. It is whose standards should govern, to whom a platform must explain its decisions, and who bears the cost when legal and political systems collide.

The United States and China have institutional differences over political expression that cannot be wished away. Trying to dissolve them in an abstract set of “global values” is unrealistic—and risks concealing the real conflicts of power beneath them.

Yet the absence of common content standards does not mean there can be no common governance procedures. The two systems may still find limited, testable agreement on notice, appeals, transparency, algorithmic audits, and responses to clearly demonstrable harm.

I. The Core Difference: Which Risk Comes First?

To understand the divide between American and Chinese speech governance, we need to examine each country’s history and institutional design. But the explanation should not collapse into a cultural slogan: America values freedom; China values order.

The American system is designed first to guard against the state using its power to suppress expression.

Memories of religious persecution in the colonial period, resistance to press controls around the time of independence, and later experiences such as McCarthyism and the Vietnam War protests all reinforced suspicion of government interference in speech. The First Amendment consequently gives political expression—and criticism of public officials—exceptionally strong protection.

Even when false statements are involved, American law weighs the chilling effect that liability could have on political criticism. Public officials pursuing defamation claims, for example, generally face the high burden of showing “actual malice.” This does not make every false or malicious statement lawful. It means that officials cannot easily use the law to silence their critics.

Nor is American public discourse free from control. The First Amendment primarily restrains the government; it does not require private platforms to remain neutral. Account suspensions, reduced distribution, demonetization, and copyright complaints can all determine what people are able to see.

The United States therefore has content governance, but much of its power has migrated from the state to private platforms. That creates a different question: when a company effectively controls part of the public square, what procedural obligations should it bear?

China’s system is designed first to guard against the risks that information flows may pose to political stability, public order, and social security.

Within this framework, the state is understood as the guarantor of order, and platforms are expected to identify and contain risks proactively. Rather than waiting for harm to occur and assigning liability afterward, Chinese governance places greater weight on early intervention. Regulation therefore tends to arrive sooner, and platforms carry broader duties to manage content.

This ordering of priorities is connected to China’s history of political and social upheaval, the risk consciousness created by rapid development, and the expansive role the state plays in social governance. But explaining why a system emerged is not the same as defending every practice carried out in its name. By the same token, defending freedom of expression does not require ignoring the harm that expression can cause.

The deeper divide is over which error each system fears most.

The American system is more concerned that power will suppress speech that should have remained free. The Chinese system is more concerned that speech that should have been stopped will be allowed to create disorder. The former is more willing to bear the cost of harmful speech left online; the latter is more willing to bear the cost of broader intervention.

Neither side necessarily fails to understand the other’s reasoning. They assign different weights to different kinds of error.

II. Political Speech: The Institutional Boundary That Cannot Be Avoided

In the United States, political satire, harsh criticism, cartoons, and fierce attacks on public officials are generally treated as part of public debate. Protecting political speech that is offensive or unsettling is considered an essential restraint on state power.

That protection is not limitless. True threats, incitement under specific conditions, defamation, harassment, and speech directly connected to criminal conduct may still be restricted. The central dispute is often not whether intervention is ever permissible, but how high a burden the government must meet before it acts.

In China, political expression is regulated more broadly and more preemptively. Political security, social stability, and public order can justify intervention before harm occurs, not only punishment afterward. Platforms are therefore expected to identify sensitive material proactively and assume responsibility for how it spreads.

No carefully softened vocabulary can erase this divide. It reflects different understandings of the state, the boundaries of individual rights, and the sources of political legitimacy.

A global platform cannot resolve the conflict with a single worldwide list of prohibited content. The same post may be protected political criticism in one jurisdiction and restricted as a public-order risk in another.

Acknowledging that fact is more honest than declaring that the world merely needs a common set of values.

III. Shared Risks Do Not Produce Shared Standards

Beyond political speech, China and the United States do face many of the same digital harms: financial scams, impersonation, child sexual abuse material, dangerous product advertising, doxxing, cyberbullying, and credible threats of suicide or violence.

Because the real-world damage is easier to identify, cooperation is more plausible here than it is in the domain of political expression.

But facing the same risks does not mean the two societies already share the same standards. Even in child safety, they may disagree over age verification, parental authority, personal privacy, and the scope of platform responsibility. Even with health misinformation, scientific judgments can change as evidence develops and can become entangled with public-policy disputes.

The better claim, then, is not that nonpolitical issues are free of value conflict. It is this:

Some concrete harms are easier to recognize across systems, making them a possible starting point for limited cooperation.

Cooperation does not require complete agreement on abstract content categories. Platforms can begin with specific targets: confirmed scam accounts, malware links, counterfeit-drug advertisements, child sexual abuse material, and violent threats with an identifiable target and credible urgency.

The more concrete the standard, the greater the chance of cooperation across systems. The more abstract the category, the more quickly it returns to political and moral conflict.

IV. The Minimum Consensus Is Not What to Remove, but How to Govern

When countries cannot agree on content standards, procedural rules offer a more realistic point of departure.

Procedural consensus does not require governments to accept the same boundaries of expression. It requires platforms to explain how their governing power is exercised. A minimum framework should include six elements.

First, notice. When a post or account is restricted, the platform should state whether it acted under local law, its own rules, or an algorithmic risk assessment. A vague reference to “community standards” is not a meaningful explanation.

Second, appeal. Users should have a way to submit evidence and obtain review. A platform may act provisionally when danger is urgent, but it should still offer retrospective review and correction.

Third, consistency. Similar cases should be treated under similar rules. Enforcement should not shift arbitrarily with a user’s prominence, commercial relationships, or political pressure.

Fourth, transparency. Platforms should separately disclose government removal demands, takedowns under their own policies, and material that remains online but is excluded from recommendation. Removal, reduced distribution, and demonetization are different interventions and should not be collapsed into one number.

Fifth, independent auditing. An audit should measure more than the volume of content removed. It should examine false positives, false negatives, successful appeals, and whether an algorithm creates systematic disparities across languages, regions, or groups.

Sixth, proportionality. Removal should not be the only tool. Depending on the level of risk, a platform might add context, impose an age gate, suspend monetization, pause recommendation, limit forwarding, freeze an account temporarily, or remove the material entirely. The more severe the intervention, the stronger the explanation and remedy it should require.

These procedures will not eliminate disagreement over content. They can, however, reduce the space for arbitrary governance.

V. Technical Cooperation: Share Evidence, Not a Definition of Truth

Fact-checking offers room for cooperation, but a single “truth database” endorsed by different governments is neither realistic nor desirable.

Who selects the evidence? How should changing scientific conclusions be handled? Where does a factual judgment end and a value judgment begin? These questions can themselves become political.

A more practical approach is to share the infrastructure of verification without demanding a common verdict. Platforms could:

• adopt compatible formats for sources and citations;
• record when a claim first appeared, changed, and spread;
• share indicators linked to confirmed scam sites and malicious accounts;
• promote provenance tools for images, audio, and video;
• make AI-generated or materially altered media identifiable; and
• disclose the strength and uncertainty of evidence when a judgment remains contested.

Countries and companies could still apply their own rules to the resulting information. At a minimum, however, they would show where the evidence came from, how the judgment was reached, and which parts remain uncertain.

The purpose of technical cooperation is not to decide the truth collectively. It is to make evidence more verifiable.

VI. Borrow the Tools, Not the Political Standards

Different systems can still learn from one another, but the most useful borrowing concerns the tools of governance, not the boundaries of political expression.

Systems that prioritize rapid response need stronger notice, appeals, and retrospective review. Temporary measures should not quietly become permanent, and platforms should not over-remove merely to minimize their own liability.

Systems that prioritize freedom of expression and procedural protection must still confront the speed at which platforms can amplify real-world harm. Procedure cannot become an excuse for delaying action against an active scam, a sustained campaign of harassment, or a credible threat of violence.

Rapid intervention and due process are not inherently incompatible. A sound system should be able to act during an emergency, reconsider once the danger has passed, and provide a genuine remedy when it gets the decision wrong.

Conclusion

Global platform governance does not have to wait for the United States and China to resolve their disagreement over values. Their institutional divide over political expression has deep historical and practical roots and is unlikely to disappear soon.

A more realistic goal is to place minimum procedural constraints on content governance in both systems: rules should be visible, decisions explainable, errors appealable, power auditable, and intervention proportionate to demonstrable risk.

This consensus would not persuade either side to adopt the other’s political standards. Nor would it resolve the enduring tension between freedom and stability. What it could do is reduce the risk of arbitrary platform power and users left without remedy when value conflicts cannot be settled.

The most viable form of cooperation is not to decide together what is true. It is to constrain together how governing power is used.

The step toward the other side need not be a step in political values. It can be a step in procedure: systems that prioritize stability can give power more explanation and accountability; systems that prioritize freedom can take greater responsibility for demonstrable real-world harm.

It is a small step. But it may be the most realistic place to begin.