The Economic Security Program of the Research Institute for Democracy, Society and Emerging Technology (DSET) released a report in June 2026 titled “The Great Breakout: Advanced Packaging and China’s Race for AI Compute Parity.” The report analyzes how advanced packaging is reshaping the global semiconductor value chain and becoming a key breakout pathway for China to compensate for its constraints in advanced-node manufacturing and continue advancing its artificial intelligence (AI) compute capabilities under the pressure of U.S. export controls.

As Moore’s Law approaches its physical limits, the competitive dimension of the semiconductor industry is shifting from front-end transistor scaling to back-end system-level integration. By integrating chiplet stacking architectures, heterogeneous integration, substrates, and high-density interconnect technologies, advanced packaging can significantly improve system performance, energy efficiency, and data transmission capacity. It has therefore emerged as a core strategic technology in the competition over AI compute.

Core finding: The U.S. export control regime has long followed a logic of “node-centrism,” focusing on advanced manufacturing equipment, electronic design automation (EDA) tools, and performance thresholds for high-end AI chips. This regime has indeed constrained China’s access to the most advanced semiconductor technologies. Yet it has also created a structural blind spot: when China cannot catch up at the frontier node on a single chip, it can still use advanced packaging to integrate multiple domestic dies that represent the best available within its process capabilities, achieving “good-enough” performance at the system level. Huawei’s Ascend series is a representative example of how system-level performance can be used to compensate for gaps at the process-node level.

The report argues that China is not simply seeking to compete head-to-head with the United States and its allies across every layer of technology. Instead, it is attempting to leverage its industrial foundations in mature-node manufacturing, outsourced semiconductor assembly and test (OSAT) capacity, substrates, printed circuit boards (PCBs), fan-out panel-level packaging, and domestic market demand to assemble “good-enough” compute capabilities sufficient to support domestic AI training and inference needs. This does not mean that China has achieved true parity in frontier AI chips. But its deployable AI compute capacity is improving, buying China time to continue advancing AI development under export control pressure.

The report argues that export controls must move beyond a chip-by-chip framework and extend to the advanced packaging supply chain. To remain effective, controls must close gaps across four layers: materials, equipment and process technologies, finished products, and anti-circumvention enforcement.

Drawing on its analysis of technology pathways, industrial policy, capability validation, and critical supply chain nodes, DSET identifies the following key findings:

  1. Paradigm Shift and the Node-Centric Blind Spot: Advanced packaging is shifting the center of semiconductor competition from transistor scaling to system-level integration. Controls centered on advanced nodes have not adequately addressed China’s buildup in the back-end supply chain. They also overlook the fact that advanced packaging and heterogeneous integration can generate meaningful system-level gains even without access to frontier nodes.
  2. Three Packaging Pathways and Asymmetric Breakout: Advanced packaging can be divided into three pathways according to the division of labor among firms: foundry-led, foundry-OSAT collaboration, and OSAT-led. China is concentrating on the OSAT-led pathway, which is cost-feasible, has lower exposure to existing controls, and offers room for upward technological extension. This logic resembles the Warring States-era strategy of “Tian Ji’s horse race”: China is not matching the United States head-to-head in every technology domain, but is instead using industrial depth and system integration to obtain “good-enough” compute capability. However, this breakout pathway still faces ceilings. Engineering bottlenecks such as interconnect precision, warpage control, and thermal management remain physical constraints as China attempts to move from mid- and lower-tier routes toward higher-density architectures.
  3. Institutional Acceleration through a Whole-of-State System: China’s ability to turn advanced packaging from a technical possibility into deployable compute is driven by institutional acceleration through national industrial policy, the Big Fund, policy finance, government procurement, and demand from national compute infrastructure. This system reduces investment uncertainty, absorbs yield-ramp and long-cycle technology iteration risks, and helps national champions and leading firms build a back-end ecosystem. 
  4. Capability Validation through Huawei Ascend: Huawei’s Ascend series demonstrates that advanced packaging can be translated into deployable AI compute under export control pressure. From the Ascend 910C’s dual-die packaging architecture to the planned adoption of quad-die packaging in future products, Huawei’s trajectory shows that multi-chip integration has become a strategic direction for improving system-level compute under process-node constraints. Using total processing performance (TPP) and memory bandwidth as two analytical axes, the report examines the performance trajectory of future Ascend products and finds that their post-packaging system-level capabilities are challenging existing control thresholds.
  5. Critical Supply Chain Nodes and Control Gaps: The report disaggregates the advanced packaging supply chain layer by layer and finds that China’s most vulnerable points are precisely where current export controls have the weakest coverage. Upstream materials such as ABF film, BT resin, and low-dielectric specialty glass fiber cloth; key components such as ABF substrates; and back-end equipment such as laser drilling, grinding, and automated inspection tools are largely not directly controlled. These gaps create potential pathways for China to obtain uncontrolled inputs and further integrate domestic dies.

The report argues that China’s advanced packaging strategy should be understood as a “great breakout.” China is not simply trying to catch up within the same process-node race. Instead, it is attempting to shift the terrain of competition: from front-end nodes to back-end interconnects, from single-chip performance to system-level integration, and from process parity to “good-enough” deployable compute capability. This pathway is especially relevant for AI inference, cloud services, telecommunications, automotive systems, and industrial applications, where not all use cases require the most advanced chips and where scaled deployment can generate both strategic and commercial value.

For the United States, Taiwan, Japan, South Korea, the Netherlands, and other democratic technology allies and partners, this means that export controls must move beyond traditional node-centric thinking. Policies to promote allied advanced packaging capacity remain important, but investment and capacity expansion alone cannot offset China’s advantages in back-end manufacturing and industrial scale. Democratic countries must also strengthen protective measures and build a more comprehensive regulatory architecture targeting China’s use of advanced packaging for technological workarounds and system-level compute expansion.

The report offers the following policy recommendations. First, export controls should be systematically expanded to the advanced packaging supply chain, covering the materials, equipment and process technologies, finished products, and system integration capabilities that support AI and high-performance computing packaging. Second, specification-based controls should be established for key upstream materials such as ABF film, BT resin, low-dielectric specialty glass fiber cloth, and advanced bonding materials, distinguishing general commercial applications from high-end AI packaging uses through technical thresholds. Third, quantified control thresholds should be established for hybrid bonding, 2.5D and 3D heterogeneous integration, high-density substrate fabrication, fine-pitch micro-bumps, and related equipment, with deeper multilateral coordination among the United States, Japan, the Netherlands, South Korea, and other allies. Fourth, controls should expand from single-chip specifications to post-packaging system-level performance, covering finished substrates, packaged AI and high-performance computing modules, high-bandwidth memory, and end-use tracing. Fifth, an anti-circumvention architecture should be built to address third-country transshipment risks through mandatory disclosure obligations, anomalous order screening, transshipment verification, and multilateral information-sharing mechanisms, with stronger enforcement cooperation focused on Southeast Asia and other high-risk regions.

In sum, DSET finds that advanced packaging has become one of the most important, yet still underappreciated, strategic arenas in the AI compute competition. U.S. export controls have limited China’s access to frontier chips, but China is using advanced packaging to raise the floor of its domestically deployable compute capacity and buy time for further technological catch-up. If the United States and its allies and partners want to preserve strategic leverage, they must upgrade export controls from a node-centric framework to a comprehensive economic security strategy covering materials, equipment and process technologies, packaging modules, system integration, and third-country enforcement.