In August, the U.S. Justice Department and the FBI dismantled the online infrastructure behind two hacking platforms operated by a Chinese state-sponsored group. It had been targeting sensitive U.S. government networks since at least 2018. In October, the agencies seized hacking tools used by another such group whose operations the U.S. government had previously disrupted in 2024.

Washington can deny Beijing’s hackers access to harmful capabilities. But the longevity and resurgence of the hackers’ campaigns raise a hard truth: Disruption alone cannot reverse the strategic advantages generated by years of Chinese cyber activity.

The conventional U.S. approach to cyber deterrence has treated Chinese operations as individual security incidents. But to deter China from accruing advantages through cyber activity, the United States must change the incentives of a vast hacking ecosystem built around the Chinese state in order to counteract its impacts on the United States’ military, technology, and trade.

A shift is now underway in Washington. In March, the U.S. government’s 2026 Cyber Strategy promised sustained and aggressive action through public-private partnerships, which would combine offensive and defensive capabilities with economic and legal countermeasures. Recent reports from the Council on Foreign Relations (CFR) and Britain’s Royal United Services Institute similarly advocate for coordinating cyber operations with other instruments of state power.

Such proposals appear to recognize that disrupting individual operations has not changed China’s broader cyber behavior. But while combining instruments may help disrupt individual operations and parts of the ecosystem, they may still fail to credibly deter the broader pattern of activity that generates cumulative advantage.

Therein lies the more troubling possibility: that the United States cannot credibly deter the broad pattern of Chinese cyber activity at a price that it is prepared—or willing—to sustain.

China’s recent cyber campaigns have advantages beyond individual intrusions. In 2023, for example, the Volt Typhoon campaign secured access to U.S. infrastructure in Guam that would be critical for U.S. operations in a future Taiwan contingency. While the United States removed some of the original footholds, the Justice Department has warned that the target infrastructure is still vulnerable; in early 2025, security analysts found that a Volt Typhoon-linked group had steadily penetrated U.S. utilities through that year. Disrupting individual intrusions had not prevented China from prepositioning for future disruption.

The year after the Guam intrusion, a campaign attributed to the Salt Typhoon hacker network compromised U.S. telecommunications networks, allowing Chinese actors to collect sensitive information from surveillance systems. More than a year after the intrusion was first disclosed, the campaign was discovered to be surveilling congressional staffers’ communications. It is still unclear whether Salt Typhoon infrastructure has been removed from the affected networks or whether it continues to accumulate intelligence.

A broad campaign of deterrence has proved to be largely ineffective, but it was once possible to think in narrower terms. In 2015, then-U.S. President Barack Obama and Chinese President Xi Jinping agreed that neither side would conduct or knowingly support cyber-enabled theft of intellectual property for commercial advantage. Security firms subsequently reported a decline in detected Chinese intrusions against U.S. companies (although the decline had begun before the agreement, and its causes remain contested).

The agreement did not end Chinese cyber espionage. It did, however, identify a relatively narrow category of conduct—commercial theft—that the two governments could explicitly agree to restrain, as opposed to the broader range of purposes that Chinese cyber operations now serve, from espionage to conflict prepositioning.

Washington no longer faces a discrete category of conduct around which to organize a deterrence effort. Today, it must determine which activities deserve a response, how much future risk they pose, and what consequences might discourage further operations. A response calibrated to an intrusion’s immediate effects may miss its future significance. Punishing every instance of access that might someday be used disruptively, meanwhile, would be difficult to sustain.

Similarly, punishing any single actor or intermediary that enables Chinese cyber operations is no longer enough to deter the wider ecosystem. Contractors disappear and reemerge frequently. The company disrupted in October provided hacking services to the People’s Liberation Army and the Ministry of State Security (MSS); others have conducted intrusions for both the MSS and the Ministry of Public Security. Rather than relying on single, self-contained operations, Chinese security organizations draw on overlapping external providers, creating an ecosystem that can regenerate operational capacity even amid disruptions.

For Washington, sanctions, export restrictions, criminal investigations, and public exposure could threaten the commercial interests and personal security of these actors. Rewards and protections encouraging people inside the enabling network to disclose information or defect, as proposed by the CFR report, could change incentives within the ecosystem, not merely punish its visible members. Pressure could make providers more cautious, expose links between companies and state agencies, and make operations harder to mount. At the state level, a credible threat of coordinated economic and diplomatic consequences might also discourage activities that Beijing considers too costly.

Yet such proposals contain significant shortcomings. Many enablers operate inside China, where they have little exposure to U.S. markets or financial systems, and their main sources of income come from state customers. In the past, sanctions have dismantled individual companies fronting the state. They would continue to have little lasting effect on companies that are dependent on Chinese security agencies, since Beijing can fund, protect, or replace providers when their work serves a strategic purpose.

Crucially, imposing costs on a contractor changes Beijing’s calculation only if the contractor’s financial, operational, or reputational losses degrade a longer-term strategic capability, threatens to expose other state-sponsored operations, or makes replacement offensive capabilities more costly for the wider ecosystem. If a state agency can absorb the loss and find another supplier, then Washington will have spent resources on punishing a node without changing the demand for its services. Previous attempts to identify and disrupt such valuable nodes have simply resulted in a temporary lull before resurgence.

To put meaningful pressure on Beijing, Washington would need to credibly signal that specific Chinese cyber activities would trigger costs substantial enough to change Chinese leaders’ calculations. That would require coordinated measures in more visible areas of strategic competition, such as restrictions on semiconductor exports or access to U.S. markets.

But China could retaliate in areas of competition where it retains leverage, such as by further restricting U.S. access to rare earth and critical mineral supplies, disrupting U.S. defense production. In that case, efforts to protect national security by deterring Chinese cyber activity could ultimately undermine it elsewhere.

Washington would thus need to accept immediate economic costs in return for uncertain deterrent effects for as long as necessary, when even sustained pressure may not stop the wider accumulation of access and intelligence. For the United States, China’s cyber operations may simply not hold enough strategic or political significance to warrant that economic price.

That leaves Washington with a dilemma. When espionage today can become disruption tomorrow, the value of deterring a wider range of cyber activity must be weighed against the costs of sustaining those restrictions and absorbing countermeasures. The challenge is to cast a wide enough net to address the sources of future risk while keeping threats specific and credible enough to influence behavior.

Accepting these risks does not have to mean abandoning deterrence; Washington can still seek to prevent China from using its access to U.S. infrastructure to disrupt essential services or pressure U.S. leaders during a crisis. But it should not confuse deterring a specific act with deterring the wider pattern of activity that accumulates strategic advantages over time.

Access may be the most valuable when it remains concealed, preserving China’s choice of whether and when to leverage it and leaving Washington to bear the costs of preparing for multiple possible outcomes. The more that Beijing believes Washington will tolerate continued accumulation, the greater its incentive to preserve that position and scale it—as the price for the United States only grows higher with time.