On August 15, 2026, the US Court of Appeals for the District of Columbia Circuit ordered a rehearing in the case of DJI's inclusion on the Department of Defense's Chinese Military Company (CMC) list. The decision is not a reversal. It is a procedural recalibration. The court ruled that the lower court's reliance on publicly available information to affirm DJI's ties to China's defense industrial base was insufficient. Now, the lower court may review classified documents. This is a signal. The Pentagon likely possesses non-public intelligence linking DJI to PLA operations. Logic remains; sentiment fades.
Context: The CMC List and Drone Supply Chain Warfare
The CMC list, established under Section 1260H of the FY2021 National Defense Authorization Act, identifies Chinese companies operating in the US that are allegedly tied to the People's Liberation Army. Inclusion does not trigger immediate sanctions. But it carries severe reputational damage and restricts DoD procurement. DJI, a Shenzhen-based private company, holds 70-80% of the global consumer drone market. Its products are used by US law enforcement, fire departments, and agricultural firms. The US military itself banned DJI procurement in 2020 under the FY2020 NDAA, but the ban did not fully eliminate existing DJI units in service. The 2024 NDAA extended restrictions to all Chinese drones, including commercial models, and mandated a security risk assessment. The CMC list is the legal backbone for this phased exclusion.
The rehearing occurs against the backdrop of the Russia-Ukraine war, where both sides use DJI Mavic drones extensively for reconnaissance and light bombing. This battlefield reality gives the US a tangible security narrative: Chinese commercial drones are now weapons of war. The Pentagon argues that DJI's data links could be exploited by China to monitor US military movements. The court's decision to allow classified evidence suggests the DoD has something more concrete than public speculation. Trust no one; verify everything.
Core: Code-Level Analysis of the US Legal Argument
From a forensic security perspective, the US case against DJI rests on three pillars: supply chain ties, data exfiltration risk, and battlefield usage. Let's examine each through the lens of technical integrity.
First, supply chain ties. The DoD claims DJI is part of China's 'civil-military fusion' strategy. DJI uses Chinese-made chips, batteries, and motors. Its supply chain is concentrated in Shenzhen, which is also a hub for PLA research. But correlation is not causation. Many non-Chinese tech companies, including Apple, have supply chains in China. The difference is that China's defense industrial base explicitly leverages civilian technologies. In 2020, DJI was added to the US Commerce Department's Entity List, restricting US exports of components. DJI responded by shifting to domestic chips like Allwinner and Rockchip. This is a classic case of forced localization. The US now wants to close the loophole by using the CMC list to block market access, not just component supply.
Second, data exfiltration risk. The US Department of Homeland Security and FAA have warned that DJI drones may transmit sensitive data to Chinese servers. DJI's SDK and transmission protocols are proprietary. Independent security researchers have published audits of DJI's firmware. I have personally reviewed one such audit from 2022. The analysis found no hardcoded backdoors, but it did reveal that DJI's cloud service stores flight logs on servers in mainland China. This is a metadata integrity issue. The data itself—coordinates, timestamps, video feeds—could be accessed by Chinese authorities under the country's national security laws. Even if DJI does not actively cooperate, the legal framework allows compelled disclosure. From a security architecture perspective, this is a legitimate concern. But it is not unique to DJI. Any IoT device with cloud storage faces similar risks. The US argument is selective. Vulnerabilities hide in plain sight.
Third, battlefield usage. The DoD's classified evidence likely includes intelligence from the Russia-Ukraine theater. Ukrainian forces use DJI drones extensively. Russian forces also use them, often purchased via third parties. The US may have intercepted communications or captured drones that reveal direct Chinese military involvement in data analysis or software updates. If true, that would be a smoking gun. But the US has not publicly released such evidence. The court's decision to allow classified evidence means the DoD can now present it behind closed doors. DJI's legal team will face a challenge: they cannot effectively rebut evidence they cannot see. This is a classic asymmetry in security litigation. The process becomes opaque.
The core insight is that the US legal strategy is shifting from 'public narrative' to 'classified proof'. This is a high-risk move. If the classified evidence is weak, the court may reject it, and the DoD's credibility suffers. If it is strong, DJI's global market access will be severely curtailed. The rehearing is a high-stakes audit of the DoD's claims.
Contrarian: The Real Blind Spot Is US Supply Chain Dependency
Most analyses focus on DJI's legal vulnerability. I argue the opposite: the US is exposing its own strategic weakness. The DoD's push to ban Chinese drones creates a market gap that US drone manufacturers like Skydio, Teal, and AeroVironment cannot fill. Their products cost 10 to 100 times more than DJI's equivalents. Skydio's X2D costs around $15,000. DJI's Mavic 3 Enterprise costs $2,000. Performance is comparable. The US 'Blue sUAS' program, which certifies secure drones, has only a few models, and none match DJI's price-performance ratio. The DoD's own 'Replicator Initiative' aims to deploy thousands of low-cost autonomous systems. But without DJI, the US must either subsidize domestic manufacturers or accept lower fleet sizes. This is a frictionless execution problem: the policy is ideal, but the execution fails due to market realities.
Furthermore, the rehearing may backfire politically. If the classified evidence is inconclusive, the court could side with DJI. That would undermine the entire CMC list regime. Other Chinese companies on the list, like Huawei and SMIC, would use the precedent to challenge their own designations. The US would lose a key legal tool. On the other hand, if the court upholds the blacklist, DJI will appeal to the Supreme Court. The process could drag on for years. Meanwhile, DJI's global sales outside the US continue to grow. In 2025, DJI's revenue increased by 12% year-over-year, driven by markets in Asia, Africa, and Latin America. The US market is important but not existential. The DoD's action is a surgical strike, but the patient is not dying.
Another blind spot: the US assumes that blocking DJI will prevent Chinese military drone development. It will not. China's defense industry already produces specialized military drones, such as the Wing Loong series. DJI is a civilian company. Its technology is dual-use, but the PLA does not rely on DJI for its core strike capabilities. The real threat to US dominance is the rapid commoditization of drone technology, which is driven by market forces, not by any single company. The US should focus on innovation, not litigation.
Takeaway: Vulnerability Forecast
The rehearing is a pivotal moment for the intersection of commercial technology and national security law. The outcome will set a precedent for how the US treats Chinese tech companies. If the court upholds the blacklist, expect a cascade of similar actions against other dual-use companies like Hesai (lidar) and Hikvision (surveillance). If the court rejects it, expect legislative efforts to bypass the judiciary. The real question is not whether DJI is a military company. It is whether the US can maintain a rules-based order while selectively applying security labels. In the end, metadata is fragile; code is permanent. The vulnerability is not in DJI's drone firmware. It is in the US legal framework's inability to distinguish between genuine threats and competitive protectionism. Silence is the loudest exploit.