Link to the code: Zae-Project / arkspace-core

Stanford's 2026 Space Tech Review: What the Numbers Actually Say

Each year the Stanford Emerging Technology Review (SETR) pulls together science and engineering faculty from across Stanford — the Hoover Institution, the School of Engineering, and the Institute for Human-Centered AI — to assess the state of ten frontier technology areas. The 2026 edition debuted in Washington, DC on January 28, with briefings aimed at policymakers and business leaders.

The space chapter is worth reading carefully. It is not a hype document. The authors are not selling anything. What it provides is a rigorous baseline: where the industry actually stands, what the critical structural weaknesses are, and where current trajectories lead if policy does not intervene.

The Infrastructure Everyone Depends On

The review opens with a framing that the commercial space industry often underemphasizes: space is not a sector. It is infrastructure.

GPS navigation, banking settlement systems, missile defense, internet access, and remote sensing all run on satellite infrastructure. These services are not optional features of modern economies. They are load-bearing. A GPS outage disrupts aviation, shipping, and financial trading simultaneously. A sustained interruption to satellite-based internet access in underserved regions has direct development consequences.

The practical implication is that space policy is infrastructure policy, with all the regulatory obligations and public-interest requirements that designation implies. Most jurisdictions have not made this connection formally. Space assets do not carry critical infrastructure designation in most national frameworks. That gap is one of the central policy concerns the Stanford review identifies.

The NewSpace Shift: What Changed and What It Costs

The commercial satellite industry’s transformation over the past decade has been well documented. The Stanford review provides the structural analysis. The shift from government-owned legacy systems — built to exacting specifications over long timescales at high unit costs — toward a NewSpace economy of private operators offering faster, cheaper, more rapidly iterable capabilities is now largely complete at the market level.

The satellite mass distribution tells part of the story. Most functional satellites in 2026 weigh between 100 and 1,000 kilograms, lighter than a motorcycle. The smallest operational satellites weigh under 10 kilograms, roughly the size of a loaf of bread. These are not experimental systems. They are commercial operational assets generating revenue.

The launch cost reductions that enabled this shift — SpaceX’s reusable Falcon 9 having driven per-kilogram-to-LEO costs down by a factor of 10 to 20 versus early 2010s pricing — are well-established. What the Stanford review emphasizes is that cost reduction alone did not create the current industry. What created it was the combination of lower launch costs with commercially available components (COTS electronics, standardized CubeSat buses, accessible ground software) that allowed smaller organizations to operate spacecraft viably.

The constraint on further growth is no longer launch cost. It is spectrum, orbital slots, and regulatory throughput.

Distributed Systems: The Dominant Architecture

One of the clearest trends the review identifies is the shift to distributed space systems: multiple spacecraft that interact and cooperate rather than single monolithic satellites performing all functions alone.

This is not a future development. It is the current state. Starlink is a distributed system. Kuiper is a distributed system. Planet’s imaging constellation is a distributed system. The operational logic is straightforward: distributed systems are resilient (losing one node does not fail the mission), scalable (add nodes to add capacity), and replaceable (failed nodes are replaced without mission redesign).

The review notes this trend as architecturally significant because it changes how performance, liability, and regulatory compliance are assessed. A single satellite with a specific licensed frequency assignment is straightforward to regulate. A 6,000-satellite constellation with dynamic frequency coordination, inter-satellite links, and autonomous maneuver capabilities is not.

The emerging applications the report highlights — in-space pharmaceutical and semiconductor manufacturing, lunar and asteroid mining, space-based solar power, and in-space logistics, assembly, and manufacturing (ISAM) — all follow the same distributed architecture pattern. They require multiple cooperating assets, long-duration operations, and infrastructure beyond what any single spacecraft can provide.

For ArkSpace’s orbital computing model, the distributed systems trend is directly validating. The Exocortex Constellation proposes neuromorphic processing nodes in LEO connected via optical inter-satellite links — exactly the architectural pattern the Stanford review identifies as the commercial industry’s current dominant approach.

The Sustainability Problem

The Stanford review is frank about debris. Operational constellations and the objects they leave behind are generating collision risks that produce additional debris, which creates additional collision risks. The cascade dynamics of Kessler syndrome are not a speculative future scenario in the review’s framing. They are a risk that current regulatory frameworks are not adequately addressing.

Four nations currently possess demonstrated anti-satellite (ASAT) weapons: China, Russia, India, and the United States. Each test of these systems has produced debris clouds. The 2021 Russian ASAT test against Kosmos-1408 generated over 1,500 trackable fragments and thousands of smaller untracked pieces. These objects orbit in the same altitude bands as the ISS and commercial LEO constellations.

The review’s policy analysis identifies a specific problem: the ITU and national licensing bodies can regulate new satellite deployments, but they cannot compel debris mitigation for existing objects. The debris that is already up is not subject to retroactive cleanup requirements. Active debris removal missions — Astroscale’s ELSA-M, ClearSpace’s mission contracted by ESA — are operational or in development, but they are small in scale relative to the problem.

The space debris cleanup efforts beginning in 2026 are the first meaningful operational response to this risk. The Stanford review situates them correctly: they are necessary but not sufficient. The orbital environment can only be sustained if launch rates are matched by deorbit rates, and current projections show that gap widening, not closing.

Policy Gaps and Governance Erosion

The review’s governance section is one of its more pointed assessments. International norms under the Outer Space Treaty (OST) — the foundational 1967 agreement establishing space as a global commons — are under increasing pressure from cislunar competition.

NASA’s Artemis program, China’s lunar ambitions, and commercial interest in lunar resources are all operating in a legal environment where property rights, resource extraction rights, and jurisdiction in cislunar space are ambiguous. The OST prohibits national appropriation of celestial bodies but does not clearly address commercial resource extraction. The Artemis Accords, signed by a growing coalition of nations, provide some bilateral frameworks, but are not binding treaty obligations.

The review identifies the Lunar Gateway — NASA’s planned cislunar station — as infrastructure that could anchor governance frameworks for cislunar operations in the 2030s. Whether it arrives in time to establish norms before commercial lunar activity scales is an open question.

What This Means for Orbital Computing

The Stanford review does not discuss orbital computing directly. Its scope is the space sector broadly. But its structural findings map directly onto the development trajectory ArkSpace is tracking.

The space economy, now at $415 billion, is being driven by commercial satellite services and launch. The distributed architecture dominating current constellation design is the same architecture that orbital computing requires. The policy gaps in debris management and spectrum allocation are real operational risks for any future orbital infrastructure.

The review’s most useful contribution is its framing of space as critical infrastructure that most jurisdictions are treating as a commercial market. The governance response appropriate to critical infrastructure — mandatory standards, liability frameworks, public-interest obligations — is not yet in place. That creates both risk and opportunity for the organizations building orbital systems in the current window.

Path Forward

The SETR 2026 space chapter is available as a PDF from Stanford and will be updated in the 2027 edition. The Hoover Institution has published excerpts and briefing summaries for the policy community.

What the review makes clear is that 2026 is not a quiet year in space. It is a year when the commercial infrastructure built over the previous decade is running at operational scale, when the policy frameworks designed for an earlier era are visibly insufficient, and when the technical capabilities — AI-managed constellations, orbital computing trials, debris removal operations — are moving from demonstration to operations.

The question the Stanford review implicitly asks is whether governance will catch up before the gaps become irreversible. The technical community’s answer, for now, is to build carefully and document the risks clearly.

Official Sources