The very large GJ-X stealth drone, a ‘cranked kite’ flying wing, is reported to have flown near Malan. Technical analysis, possible missions and strategic implications.
Summary
China is reportedly conducting flight tests on a very large stealth drone, nicknamed GJ-X, with a cranked kite wing design. The aircraft was spotted in September at the Malan test site (Xinjiang) via satellite imagery, with an estimated wingspan of approximately 42 m. A short clip released in mid-October shows an aircraft of similar configuration in flight, with split rudders and exhaust bumps suggesting a twin-engine design. The underside appears to feature counter-shaded paintwork to break up the visual signature. Its exact role remains under debate: UCAV with attack capability, ‘unmanned bomber’ or large penetrating ISR platform, similar to the American RQ-180. In any case, the GJ-X would be part of a sustained stream of Chinese programmes (J-36, J-XDS, GJ-11) aimed at achieving information superiority and remote air power. The challenges are clear: inter-theatre range, modular payload (sensors, weapons, decoys), endurance above 12,000 m, and multi-spectral survivability (shape, materials, thermal management). At this scale, the aircraft falls into a very rare class of stealth drones, below that of B-21 bombers, but well beyond X-47B-type demonstrators.
The programme and configuration: a large-scale ‘cranked kite’ flying wing
The most tangible element is the repeated observations made at the Malan test site. In September, a very dark aircraft, positioned at the runway threshold, was observed to have a wingspan estimated at ~42 m (137 ft) with a cranked kite leading edge and a fuselage embedded in the wing. This geometry optimises lift at high aspect ratio while managing stability and internal payload constraints. In flight, split rudders on the upper surface/wings suggest solutions already seen on the B-2 for yaw control without a tailplane, limiting vertical surfaces and therefore radar drag. A bulge above the rear suggests a recessed double flow, hinting at two integrated turbojets, with S-shaped ducts and concealed exhaust lips to reduce IR and dorsal SER.
The twin-engine option is consistent with the estimated weight (in the order of several dozen tonnes at take-off), redundancy at high-altitude cruising and the ability to simultaneously carry heavy sensors (large aperture AESA, ELINT/COMINT suites, side antennas) and cargo holds. The dark-contrast ventral paintwork, depicting fictitious ‘wings’ and a ‘fuselage’, is a form of counter-shading designed to disrupt visual identification from certain angles and in certain lighting conditions, which is useful above 10,000–15,000 m.
In terms of performance, a 42 m wing with reduced relative thickness and moderate sweep suggests a cruising Mach of around 0.7–0.8, a ceiling > 15,000 m and an endurance > 20 hours with structural tanks. As a benchmark, an X-47B (18.9 m wingspan) has a maximum weight of ~20 t and a much lower endurance; conversely, an RQ-180 is credited with a 24-hour endurance and a wingspan estimated to be > 39–40 m. The GJ-X would therefore occupy a niche as a strategic flying wing capable of carrying several tonnes of internal payload (sensors, ammunition, decoy supplies), while maintaining low observability across a wide spectrum thanks to its profile, the serrated edges of its hatches and the treatment of its exhaust. The air intakes, which are not visible in the photos, suggest either low dorsal intakes with radar covers or deeper ‘serpentine’ integration at the inner leading edge. In any case, the design aims to reduce echoes in the front and side quadrants, which is crucial for penetrating modern A2/AD bubbles.
Operational role: UCAV, unmanned bomber or large ISR platform
The doctrinal debate focuses on its use: kinetic strike, penetrating reconnaissance or versatility. A UCAV of this size would enable deep raids with internal bays for guided weapons weighing 250 to 1,000 kg, high-altitude penetration profiles followed by terminal descent, or stand-off release (subsonic air-to-surface missiles > 250 km). The realistic total payload (around 2–6 t depending on configuration and range) would also allow for jamming kits, active deception (DRFM), towed or consumable decoy dispensers, and large-diameter E/O-IR pods.
In the ISR variant, the wing area and central volume facilitate the integration of a synthetic aperture AESA radar (X/Ku band) in a bar configuration, passive antennas (ELINT/COMINT) around the entire perimeter, and a long-focus ventral optronic mast. Typical mission profile: transit > 1,500 km, orbit at 15,000–18,000 m for 10–18 hours, metric resolution SAR scanning, GMTI mapping, transmitter geolocation, real-time correlation via broadband links. Such penetrating ISR fills the gap between satellites (slow revisits, orbital constraints) and manned platforms (cost/political risk).
A versatile option remains credible: the airframe would perform sequential missions (ISR targeting, then stand-off firing) by imposing strict emission discipline. The split rudders and absence of tail fins confirm the emphasis on directional stealth and long-range cruising capability. Finally, cooperation with smaller swarms (GJ-11, loyal wingmen) would provide saturation and role distribution: the GJ-X ‘carrier’ hosts sensors and relays, while the escort elements perform jamming and strikes. In a Taiwan-East China scenario, such a system would create sensor persistence beyond the first island chain, with a high mission cadence if two or three cells rotate in a ‘carousel’ (take-off every 6–8 hours), increasing ISR pressure and dynamic targeting capability (multi-domain kill chain).
Industrial positioning: comparisons and technical readings
Compared to GJ-11 (Sharp Sword), a more compact flying wing (~14 m wingspan), the GJ-X clearly belongs to a higher class: strategic range, deeper bays, and larger sensors. Compared to the X-47B, a naval demonstrator, the GJ-X is less focused on landing and more on sustained penetration into contested space. The most relevant analogy remains the American RQ-180, an unarmed HALE stealth ISR platform according to open sources: large wingspan, 24-hour endurance, silent penetration, multi-spectral collection, data relay. The GJ-X could combine this ISR logic with a kinetic payload capacity, an option that the United States has tended to reserve for other vectors (B-21, air-to-air/air-to-ground missiles launched from manned platforms).
Aerodynamically, the ‘cranked kite’ optimises load distribution (high-lift internal sections, external panels with adapted incidence) and manages unsteady interactions at high incidence. The split rudders produce a yaw moment without penalising the SER as a fin would. The choice of a twin-engine design suggests a focus on altitude/range, but also on availability for long missions. In terms of materials, we expect a carbon-epoxy composite skin, broadband absorbent coatings, sawtooth joints, and heat treatments on the exhaust to mask the IR plume up to 3–5 µm. The antennas may be conformal (AESA in the wall), which explains the apparent absence of protruding radomes.
Economically, a 40+ m wingspan airframe, strategic sensors and limited production line would most likely cost in excess of €150–250 million per unit, extrapolating from comparable foreign programmes (indicative costs not public). The hourly cost would be significantly higher than that of a MALE (MQ-9-like) and closer to that of specialised HALE platforms, but offset by capability effects: reduced human risk, sustained penetration, exploitation of defence blind spots. China would thus have a strategic surveillance and stand-off strike tool capable of reshaping the KPPs (Key Performance Parameters) of its air operations beyond 1,500–2,500 km.
Strategic consequences: deterrence, A2/AD and operational tempo
An operational GJ-X would change three variables. Firstly, ISR penetration: persistent observation of enemy air bases, logistics hubs and naval groups, with the production of near real-time targeting orders. In a Western Pacific theatre, endurance > 20 hours and ceiling > 15,000 m allow for orbit profiles beyond 1,500 km from the coast, while retaining the ability to approach for a refined track. Secondly, the remote mass effect: combined with loyal wingmen and expendable swarms, the GJ-X would serve as a ‘mother node’ to coordinate jamming, decoys and strikes, saturating multi-layered defences (S/X radars, long-range SAMs, AEW-guided fighters). Third, tempo pressure: ISR persistence coupled with stand-off weapons (300–1,000 km) shortens the detect-decide-strike loop.
For adversaries, the response involves: more passive radars (detection of low RF/IR signatures), more high-altitude interceptors with IRST sensors, and better distributed sensor networks. On the naval defence side, the threat of a large stealth UCAV requires more remote CAPs and AWACS aircraft carried by carrier strike groups. The costs follow: a credible A2/AD system against this type of platform requires multi-static sensors, more sensitive IR/active homing missiles, and robust data fusion networks.
Finally, on the global market, the GJ-X showcase consolidates China’s perceived lead in unmanned stealth flying wings: beyond exportable MALE drones, Beijing is signalling its mastery of penetrating HALE platforms, which are difficult to export but are being showcased for their political and military value. The technological gap is becoming a message. Western budgets (led by the United States) have already acknowledged the rise of swarms and ‘quarterback’ drones; the GJ-X is accelerating this trajectory and pushing for funding for passive countermeasures, network resilience and dedicated high-altitude interceptors.
Live a unique fighter jet experience