Study
Not Out of Control
Why China Tolerates Irresponsible Behavior In Its Cyber Campaigns – And How To Respond
Executive Summary
The PRC’s cyber campaigns are widely read as an ecosystem that the government in Beijing no longer fully directs. A senior FBI official has said publicly that the Chinese “hacker-for-hire ecosystem has gotten out of control”, and indictments have called it reckless and indiscriminate. The evidence supports a different reading. The central government has not lost control of its cyber campaigns. It tolerates the looseness they produce, as the price of operating at scale. This matters for affected states: a government that tolerates looseness is likely to change course only when keeping it costs more than dropping it.
A system built for scale. Over more than a decade, the PRC put in place the laws, institutions, force structure, and talent base needed to run many cyber campaigns in parallel. It also drew in a commercial market in access, infrastructure, and tooling that grew largely on its own. Reporting duties send newly found vulnerabilities to the state, which stockpiles them for selective use and also buys exploits for operations at scale. Beijing controls the security bureaus, but tasking is spread across the provinces and part of the work is contracted out. AI is being built into the intrusion chain. Dutch military intelligence assesses that the PRC’s offensive cyber capabilities are now probably on par with those of the United States.
Control is bounded, not lost. The central government lets its operators work loosely, within a lower limit it enforces and an upper limit that has so far held. Between the two, contractors compromise targets on their own initiative and sell the results to state buyers, espionage blends with financially motivated crime, and tools and exploits spread after use. The central government corrects none of this. Since 2021, more than twenty enforcement actions have suspended leading firms from military procurement or excluded them from it. They all concerned how the firms won contracts. None concerned how they operate. Enforcement against cyber-enabled fraud shows that the state can correct at scale, and the criminal and contractual instruments that could reach operators have recently been strengthened. Correction reached how operators operate only once: around the 2015 US–China agreement, after sustained pressure on the relationship as a whole. The effect did not last. A state that can correct and does not has made a choice.
The upper limit – the ceiling – has not been tested. No PRC campaign has been shown to aim at disrupting the operation of foreign critical infrastructure, although access to power, water, and transport systems has been held for years. Restraint is likely part of the explanation, but some of that access may not be enough for disruption, and no situation has yet arisen in which disruption would have served a Chinese objective. Access that PRC actors hold in critical infrastructure should be treated as access they may one day use to disrupt it.
Responsibility does not depend on tasking. A state that builds up the system, declines to correct how it runs and keeps buying what it produces is responsible for that output, whether or not it ordered each intrusion. Whether the state ordered a specific intrusion only shows how the campaign was run. It does not limit the state’s responsibility.
More campaigns ahead. The number of PRC cyber campaigns is likely to rise, as AI and a competitive supplier market make operations cheaper. Operators are likely to care less about being named, even as they invest in staying undetected, and the gap between disciplined elite operations and loose contractor work is likely to widen.
The gap lies in the costs. Affected states have attributed, indicted, sanctioned, and taken down infrastructure, and the PRC has absorbed it. The activity stays below the threshold at which a decisive political response becomes likely. Responses mostly address single incidents and not the pattern they form, and rely mainly on technical instruments. The economic and diplomatic channels to which the PRC has visibly reacted in the past remain largely unused for cyber conduct.
Seven measures for affected states. None requires new legal powers. Two things are still missing, though: a shared record and a response agreed in advance.
-
Default attribution comes first, and the others support it: hold the PRC responsible by default for its share of any campaign with a China nexus, without first proving that it tasked the specific intrusion.
-
Supply-layer disruption takes down infrastructure that many operators share and produces evidence under judicial authority.
-
Personal attribution names the firms, universities, individuals, and bureaus behind an intrusion set.
-
Collective cost imposition uses the accumulated record for economic and diplomatic measures. Willing states act together and agree beforehand to share the cost of Chinese counter-pressure.
-
Ceiling contingency prepares for the disruption of critical infrastructure: continuity planning for an adversary already inside, a coalition response agreed in advance, and a public statement that even holding such access will draw consequences.
-
Record consolidation turns the evidence affected states already hold into one continuous, shared record.
-
Resilience reduces what intrusions yield, pays off whatever Beijing decides, and can be started by a single state today.
What would change this assessment. The central government correcting how its operators behave in the field; a PRC campaign aimed at disrupting the operation of critical infrastructure; or a measurable change in PRC behavior after a group of affected states imposes costs together.
The PRC built up the system, tolerates how it runs, and keeps buying what it produces. The delegation is the policy, and affected states are entitled to respond to it as such.
Author
Dr. Sven Herpig
Lead Cybersecurity and Emerging Threats | Advisor to the Executive Director
Related Content
Assessing Irresponsibility in Cyber Operations
A Guide for Operators and Decision-Makers in Times of Strategic Competition
Dr. Sven Herpig
Active Cyber Defense
Toward Operational Norms
Dr. Sven Herpig
Cybersicherheit und die Volksrepublik China
Ein Überblick aus deutscher Perspektive
Dr. Sven Herpig