For better or worse, technology at heart is—except to the extent that artificial intelligence fundamentally becomes involved—not so much a creator as a facilitator and enhancer of human acts, actions and activities, allowing them to become more effective, less costly, or sometimes even just merely feasible. Perhaps nowhere that is more pertinent then when it comes to human activities in outer space, which are still overwhelmingly conducted remotely and hence crucially dependent on technology. Given that “the law” has always been geared to address humans and their acts, actions, and activities, this gives rise to a rather special approach to maintaining and further developing a legal regime for outer space. The present Article intends to address and assess some of the most pertinent aspects of the unique body of space law from precisely this perspective, to shed some light on how “the law” would, could, and/or should handle relevant human endeavours in or with regard to outer space, in particular in the context of legal responsibilities and liabilities.
Sources and Creation of Law: Treaty, Custom, Jus Cogens, General Principles
From the launch of Sputnik I in 1957 to proposals for In-Space Servicing, Assembly and Manufacturing (ISAM) and new lunar activities such as resource utilization, advancing technology has always been a driving factor in the creation of space law. From a legal-historical perspective, the notion of law as creation should be contextualized in a broader legal-philosophical transition that began with the rise of positivism. Article VI of the Outer Space Treaty orbits unsteadily between international obligations and national implementation measures, rendering significant States’ understandings of those provisions. Our understanding of Article VI turns on perhaps the most creative legal endeavor: interpretation. Bing Cheng established Article VI as a lynchpin between international obligations and national measures by finding in its first sentence an attribution clause extending responsibility to non-governmental activities falling under the jurisdiction of States. Though Cheng’s interpretation has been accepted by scholars, and some domestic rules evidence its employ by States, the interpretation has been assailed on the basis that Cheng did not follow the strictures of the Vienna Convention on the Law of Treaties (VCLT). Codification, such as the VCLT, is itself an act of creation, which can have unintended consequences. Through the lens of Article VI, this Article explores interpretation as creation. It seeks to demonstrate that antipodal interpretations can be correct, that our determination of which interpretation to follow involves something other than a strict, positivist approach, and that the outcome of this debate may be more significant than perceived as states create a path forward for new space activities.
Large Language Models (LLMs) have the potential to transform public international lawyering in at least five ways: (i) helping to identify the contents of international law; (ii) interpreting existing international law; (iii) formulating and drafting proposals for new legal instruments or negotiating positions; (iv) assessing the international legality of specific acts; and (v) collating and distilling large datasets for international courts, tribunals, and treaty bodies.
This Article uses two case studies to show how LLMs may work in international legal practice. First, it uses LLMs to identify whether particular behavioral expectations rise to the level of customary international law. In doing so, it tests LLMs’ ability to identify persistent objectors and a more egalitarian collection of state practice, as well as their proclivity to produce inaccurate answers. Second, it explores how LLMs perform in producing draft treaty texts, ranging from a U.S.-China extradition treaty to a treaty banning the use of artificial intelligence in nuclear command and control systems.
Based on these analyses, the Article identifies four roles for LLMs in international law: as collaborator, confounder, creator, or corruptor. In some cases, LLMs will be collaborators, complementing existing international lawyering by drastically improving the scope and speed with which users can assemble and analyze materials and produce new texts. At the same time, without careful prompt engineering and curation of results, LLMs may generate confounding outcomes, leading international lawyers down inaccurate or ambiguous paths. This is particularly likely when LLMs fail to accurately explain particular conclusions. Further, LLMs hold surprising potential to help to create new law by offering inventive proposals for treaty language or negotiating positions.
Most importantly, LLMs hold the potential to corrupt international law by fostering automation bias in users. That is, even where analog work by international lawyers would produce different results, humans may soon perceive LLM results to accurately reflect the contents of international law. The implications of this potential are profound. LLMs could effectively realign the contents and contours of international law based on the datasets they employ. The widespread use of LLMs may even incentivize states and others to push their desired views into those datasets to corrupt LLM outputs. Such risks and rewards lead us to conclude with a call for further empirical and theoretical research on LLMs’ potential to assist, reshape, or redefine international legal practice and scholarship.