In 2021, a Starlink satellite and OneWeb spacecraft came dangerously close to colliding in orbit. While both companies’ automated systems detected a potential collision, the incident sparked intense debate about communication and de-confliction between autonomous space systems. As competing satellite operators increasingly rely on automated collision avoidance software to manage their growing constellations, this near-miss highlights a crucial gap in international outer space law: Legal frameworks never contemplated a future where automated systems, sometimes guided by AI, would make critical decisions in outer space that could constitute “harmful interference” under existing international law — that is, when a country’s space operations intrude on the activities of other countries operating in space.
Integrating autonomy into space operations has created a fundamental mismatch between the state-based, human-centric communication requirements of existing space law and the realities of autonomous space systems operated by both state and non-state actors. As I teach cadets at the United States Air Force Academy — some of whom will become operators managing these very systems — it has become clear that the existing understanding of harmful interference in space is dangerously outdated.
To address the problem, the United States should work to modernize the legal framework governing space activities. The effort should focus on embracing instantaneous consultations between non-human actors to avoid space collisions, updating standards for harmful interference to include the potential for cascading damage in space, and amending U.S. space policy to identify when automated systems make decisions. The proliferation of AI-enabled autonomous systems in space and mounting collision-avoidance incidents in recent years demand an urgent update to international space law.
A 20th Century Legal Framework Collides With 21st Century Technology
The scope of the problem of overcrowded outer space extends far beyond individual collision avoidance scenarios. Today’s space operations increasingly rely on automated systems for everything from debris tracking and satellite navigation to power management and communication routing. Such systems make thousands of small but crucial operational decisions daily, potentially affecting the activities of other space actors. The European Space Agency employs machine learning algorithms to predict collisions and automatically adjust satellite orbits, while the Defense Advanced Research Projects Agency’s Blackjack program develops autonomous satellite constellations that can independently coordinate their decisions and maneuver without human input from the ground.
In November 2024, Oman launched its first satellite for remote sensing that relies on AI for real-time data processing. Additionally, SpaceX made tens of thousands of autonomous collision-avoidance maneuvers in the first six months of 2024. The rapid proliferation of data processing and collision avoidance systems has created an operational environment that the drafters of the key outer space treaty could not have envisioned.
Article IX of the 1967 Treaty on Principles Governing the Activities of States in the Exploration and Use of Outer Space, including the Moon and Other Celestial Bodies — or the “Outer Space Treaty” — requires that both states and non-state actors undertake “appropriate international consultations” before proceeding with any activity that may harmfully interfere with the actions of other entities. This requirement assumed that operators would have time to communicate and, if necessary, de-conflict actions that might affect others. However, the reality of modern space operations has rendered this Cold War–era assumption obsolete. When automated systems make split-second decisions to alter a satellite’s orbit to avoid collisions, there is no time for the kind of consultation envisioned by the treaty. Moreover, such autonomous maneuvers may constitute harmful interference. In 2021, China’s Tiangong space station was forced to maneuver twice to avoid potential collisions with two SpaceX Starlink satellites. A Chinese source questioned whether SpaceX deliberately sought to test China’s sensitivity to close approaches.
The presence of autonomous systems in space exposes three critical flaws in the Outer Space Treaty. First, the treaty’s consultation requirement implicitly assumes human decision-making timelines that are slower than autonomous systems. Second, the standard for what constitutes harmful interference fails to account for the complex interdependencies created by networked autonomous systems. Third, disagreements between powerful countries have precluded the treaty from being updated to account for attributing responsibility when autonomous systems interact with each other.
New Developments and Near Misses
Legal issues lead to real-world challenges when satellites with autonomous systems begin to overcrowd orbits. In March 2023, an Astroscale debris-removal demonstration showcased how autonomous systems challenge the traditional understanding of harmful interference. The mission’s AI-driven mechanism for capturing in-orbit satellites made independent decisions about approaching debris, potentially raising the alarm about liability and consent that existing space law fails to consider. Similarly, OneWeb’s deployment of autonomous orbit-raising capabilities represents a significant shift toward autonomous space operations that can affect multiple actors simultaneously.
The U.S. Air Force’s Global Futures Report predicts that by 2035 most military and commercial space operations will rely heavily on autonomous systems. This shift is already evident in programs like the Space Development Agency’s National Defense Space Architecture, which envisions a constellation of satellites capable of autonomous threat response and collaborative decision-making. These systems interact with commercial autonomous platforms, creating complex webs of autonomous and, eventually, AI-driven decision-making that far exceed the scope of current space law.
Consider the implications of the breakthrough in autonomous space system coordination by the Defense Advanced Research Projects Agency’s Blackjack Program. The program showed that AI-enabled satellites could independently adjust their orbits, modify their communication patterns, and respond to potential threats without human intervention. While this capability enhances operational resilience, it also raises questions about interference when multiple autonomous systems interact in ways that affect each other’s missions. The Astroscale and OneWeb examples reflect how private actors can match and exceed government autonomy efforts. Increased levels of autonomy, while operationally efficient, create scenarios where automated systems must balance competing priorities in ways that affect other space actors’ activities.
Proposing a New Legal Framework
To address the integration of autonomy in space, the United States can take three steps toward establishing a new legal framework governing such activity. First, the United States should establish clear parameters for communications between automated systems. Space operators require greater transparency when autonomous systems are used in orbit by others, but operators should also reveal the extent to which they rely on autonomy for their systems. The United States should enforce clear usage of autonomy in space (and potential liability) using preexisting licensing requirements. For example, the Federal Aviation Administration currently licenses all launches and reentries from space. Additionally, the Federal Communications Commission allocates the radio frequency spectrum through which U.S. operators communicate with their satellites. These gatekeeper functions authorize space activities and can take steps to enforce the responsible use of automated systems in space. This is even more vital as AI is more likely to be integrated into space systems in the future.
Second, the United States should develop an updated definition for harmful interference in space, considering the cumulative effects of autonomous systems’ interactions. This definition should include an updated framework for fault-based liability as well as the possibility of cascading effects following a collision in space. If a space operator’s automated system causes a collision, that state must be liable for immediate damage and follow-on damage from space debris. This would lead to more responsible use of autonomous space systems, putting space actors on notice and encouraging responsible use of autonomy. With greater predictability for the use of autonomy in space, other states might follow the U.S. example.
Finally, space operators would benefit from legal frameworks to attribute responsibility when automated systems make decisions that affect other operators. The first Trump administration recognized the need for such updates through Space Policy Directive-5, which called for new approaches to space system cybersecurity and automated processes. Additional measures should now be taken to regulate the reckless use of new or experimental types of autonomy in space, such as establishing a “reasonable person” standard for attributing responsibility.
In making these moves, the United States could account for the rapid proliferation of autonomous space systems operated by state and non-state actors. However, any lasting change on this issue would likely need to be made through a multilateral effort. Ideally, this should take place through a U.N. framework such as the Outer Space Treaty.
The Stakes Are High
As commercial space activities expand and more state and non-state actors deploy autonomous space systems, the risk of accident increases exponentially. The recent growth in satellite launches — with over 7,000 active satellites now in orbit — has already strained the ability to manage space traffic. Adding more autonomous systems — and ones eventually capable of decision-making through AI — to this increasingly crowded environment without updating the legal framework invites potential disaster. The need for change is clear to anyone training the next generation of space professionals.
Every year, cadets at the Air Force Academy grapple with scenarios that highlight the growing disconnect between current space law and operational realities. For instance, do anti-satellite missile tests constitute harmful interference with another space actor? While the existing legal interpretation on the issue is unclear, crowded satellite constellations in low earth orbit suggest that they do. Cadets’ insights regularly remind me that they are training to operate in a space environment governed by legal frameworks that long predate today’s technology.
The time has come to align international space law, especially the concept of harmful interference, with the era of autonomous and AI-enabled space operations. The United States should update its legal frameworks to account for autonomous systems while upholding the Outer Space Treaty’s core principles of international cooperation and mutual benefit in space activities.
Maintaining the status quo risks rendering Article IX meaningless, precisely at a time when the United States should have strong legal frameworks to manage an increasingly complex space environment. The success of commercial space ventures, the safety of military space operations, and the sustainable development of space activities all depend on the ability to adapt international law to this new reality. Continuing to operate autonomous systems under legal frameworks that never contemplated their existence risks turning space into a domain where technology outpaces the ability to govern it safely and responsibly.
Matthew Ormsbee is an assistant professor of law and the director of space law at the United States Air Force Academy. He earned his master’s degree in air and space law from McGill University in 2023. The views expressed are those of the author and do not necessarily reflect the official policy or position of the U. S. Air Force Academy, the U.S. Air Force, the Department of Defense, or the U.S. government.
Image: European Space Agency via Wikimedia Commons.
Jesus Christ is King
