Preface:

The capital markets excel at discounting distant prospects ahead of time. Even as a satellite orbits in space and a company awaits its official listing, the imagination surrounding 'Hong Kong's first AI satellite stock' has already taken flight.

Behind Guoxing Aerospace's 11.6 billion valuation lies both the capital market's pursuit of scarce assets and the risk premium associated with an incomplete commercial closed loop (closed loop).

Author | Fang Wensan

Image Source | Internet

The Allure and Value of AI Satellites

Traditional remote sensing satellites operate on a 'capture first, transmit to ground, then analyze' model.

This approach was effective in the past but has inherent bottlenecks: limited downlink bandwidth, transmission delays, ground processing queues, and a lengthy path before raw data generates value.

When satellites are equipped with AI payloads and computing capabilities, the logic shifts.

They can filter, identify, compress, and judge data in orbit, discarding invalid (invalid) data and prioritizing critical information for ground transmission.

This is the appeal of 'space-based computing,' enabling satellites to replace ground-based cloud computing by shifting edge judgment capabilities to the data source.

Just as autonomous vehicles do not transmit every road frame to the cloud for decision-making, future satellites cannot indiscriminately downlink all raw data for centralized processing.

The value of spatial computing stems from 'near-source processing,' a cornerstone of Guoxing Aerospace's narrative.

It elevates commercial space from mere hardware manufacturing to a composite system of 'data, algorithms, computing power, and applications.'

Satellites handle data collection, AI enables recognition, constellations ensure coverage, and ground platforms deliver industry-specific solutions. Each layer sparks commercial imagination but also risks becoming a cost sinkhole.

Its most capital-friendly attributes lie in 'scarcity' and 'option value.' In Hong Kong's market, few companies truly center on full-stack capabilities in AI satellites, spatial computing, and commercial space.

Scarce assets with listing expectations often see prematurely inflated valuations.

Guoxing Aerospace's prospectus repeatedly emphasizes keywords: commercial space, artificial intelligence, AI satellites, spatial computing, and space-based solutions. It aims to package satellite development, payload design, in-orbit operations, data processing, and industry applications into a cohesive capability.

This positioning is critical. In the AI era, satellites are no longer just orbital cameras and signal relays but potential in-orbit edge computing devices.

The 11.6 Billion Valuation: Scarcity, Market Trends, and Technical Barriers

Guoxing Aerospace's valuation surge is rare in the hard tech sector.

Valued at approximately 6.538 billion yuan by the end of 2023, its valuation nearly doubled in just over two years, driven by three overlapping pricing logics.

① Scarcity Premium: Among Hong Kong-listed companies, pure commercial space plays are scarce, and firms focused on AI satellites and space-based computing are virtually nonexistent.

Current Hong Kong-listed satellite-related companies include AsiaSat (traditional communication satellite leasing), Jingwei Tiandi (ground communication engineering), and Interstellar Aerospace Technology (still in early business transformation).

If listed, Guoxing Aerospace would become Hong Kong's bona fide 'first space AI stock,' a scarcity (scarce) asset supporting its valuation for funds eager to allocate to aerospace tech.

② Systemic Industry Uplift: The 2026 government work report included aerospace among six emerging pillar industries, alongside integrated circuits and biopharma.

The National Space Administration established a commercial space bureau, and over 20 provinces/municipalities have introduced targeted support policies.

CCID Think Tank data shows China's commercial space market reached 2.83 trillion yuan in 2025, expected to surpass 3.5 trillion yuan in 2026, with a five-year CAGR of 23.1%.

Elevated policy positioning directly raises sector valuations, compounded by global aerospace asset revaluation amid SpaceX's anticipated listing.

③ Technical First-Mover Barrier: Guoxing Aerospace is not merely a satellite manufacturer; its core edge lies in deeply integrating AI computing with satellite hardware.

The prospectus reveals seven generations of AI payload iterations, with the latest generation offering no less than 10 POPS of computing power.

While peers debate space-based computing feasibility, Guoxing has validated the full chain from concept to in-orbit operation, forming the technical foundation of its valuation.

Financial Growth Amid Persistent Losses

The other side of the 10-billion valuation is a financial reality yet to achieve profitability. Analyzing Guoxing Aerospace's financials over three years reveals the tension between tech investment and commercialization.

Revenue grew steadily: 508 million yuan (2023), 553 million yuan (2024), and 703 million yuan (2025), a 27% YoY increase in 2025.

The revenue mix shifted positively: satellite-related services surged from 3.22 million yuan in 2023 to 257 million yuan in 2025, rising from 0.6% to 36.5% of total revenue.

Traditional space-based solutions revenue fell to 444 million yuan, accounting for 63.2%.

The company is transitioning from a project-based solution provider to a satellite operations and computing service provider.

Widening losses drew more attention: net losses were 139 million yuan, 177 million yuan, and 256 million yuan over three years, totaling over 570 million yuan in cumulative losses.

The primary driver was surging R&D investment: 53.478 million yuan (2023), 142 million yuan (2024), and 152 million yuan (2025), totaling over 340 million yuan in three years, with R&D-to-revenue ratios peaking at 25.7%.

The prospectus reveals that the top five clients accounted for 78% of revenue, primarily local governments, state-owned enterprises, and research institutes.

This To G-dominated structure ensures order stability but also leads to high accounts receivable.

Huatai Securities' telecom industry research notes that domestic commercial space firms are generally transitioning from technical validation to scaling (scaled) applications, with revenue heavily reliant on policy-driven orders and marketization (market-driven) pay (payment) scenarios yet to fully materialize.

The Promise and Challenge of Space-Based Computing

The 'Space Computing Plan' centers on 'space-based processing,' deploying AI computing power directly on satellites for in-orbit tasks like data processing, target recognition, and change detection, transmitting only conclusions to Earth.

Guoxing's test data shows this model compresses data response times from hours to minutes and reduces transmission volume by over 90%.

For time-sensitive scenarios like emergency rescue, disaster monitoring, and low-altitude economy, this path holds clear value.

The planned constellation comprises 2,400 inference satellites and 400 training satellites across orbits, networked via inter-satellite laser links to form 100,000 POPS of inference computing and 1 million POPS of training computing.

Targeting the booming AI Agent market, autonomous vehicles, drones, and intelligent robots could access space-based computing globally.

The narrative is compelling, but commercialization faces significant hurdles.

Networking 2,800 satellites, even at a reduced per-unit cost of 1 million yuan, entails billions in hardware investment, plus launch and operational costs.

Currently, space computing clients are concentrated in government emergency response and land surveying, with limited market-driven commercial demand.

While AI Agents accessing space computing is technically validated, commercial willingness to pay, pricing models, and service standards remain exploratory.

As the space computing sector heats up, state-owned enterprises and private peers are accelerating layout (deployments).

CASC and CASIC have deep expertise in satellite platforms and payloads, while companies like Galaxy Space and Gesi Aerospace are advancing onboard computing R&D.

Internationally, SpaceX's Starlink Gen 2 offers onboard processing, and Amazon's Kuiper Project is also pursuing space computing.

Guoxing holds a first-mover advantage, but its durability remains uncertain.

Conclusion:

Behind the vibrant narrative, sobriety is warranted. Space is not a server room; satellites face unique challenges in power consumption, heat dissipation, radiation, reliability, and maintenance compared to ground data centers.

Algorithms cannot be infinitely scaled, computing power cannot be dynamically expanded, and hardware failures cannot be swiftly replaced like ground servers.

The true challenge for AI satellites is delivering usable, reliable, and sustainably operable computing power to orbit while ensuring customers keep paying.

This step will determine whether 'Hong Kong's first AI satellite stock' is a fleeting capital market halo or a gateway to commercial space's new infrastructure.

Partial References: Jiemian News: 'Guoxing Aerospace Again Aims for Hong Kong's 'First Space AI Stock' with 11.6 Billion Valuation',

CCID Think Tank: '2025-2026 China Commercial Space Industry Development White Paper', Frost & Sullivan: 'China Commercial Space Industry Research Report', Economic Observer: 'High Client Concentration Challenges Commercial Space Monetization', China Daily: 'Guoxing Aerospace Unveils Full-Chain Validation Progress for World's First Space Computing Constellation'