Technical Evolution of Lethal Autonomous Weapon Systems (LAWS) and Autonomous UAVs: A Comprehensive Analysis of International Humanitarian Law, Supply Chain Control, and Sovereign Edge Defense
Sep 2026· Zenodo (CERN European Organization for Nuclear Research)
Ethics and Social Impacts of AI
Abstract
概要 (Description / Abstract in Japanese) 本論文は、自律型致死兵器システム(Lethal Autonomous Weapon Systems: LAWS)および人工知能(AI)を統合した自律型無人航空機(UAV)・徘徊型兵器(Loitering Munitions)の急速な軍事利用に伴う技術的・法規範的・地政学的課題を多角的に分析した学術プレプリントである。 現代の紛争環境において、ドローン兵器は遠隔操縦型から、GPS途絶環境下における視覚的自己位置推定(Visual SLAM)、車載エッジAIによる自動標的認識(ATR)、複数機による分散協調制御(スウォーム知能)を備えた完全自律型兵器へと移行しつつある。この軍事的意思決定サイクル(OODAループ)の極限的加速は、国連特定通常兵器使用禁止制限条約(CCW)の政府専門家グループ(GGE)で議論される「有意義な人間による制御(Meaningful Human Control: MHC)」を形骸化させ、「デジタル脱人間化(Digital Dehumanization)」を招く重大な懸念をもたらしている。 本研究では、オープンソース化されたAIモデルの拡散をソフトウェア単体で統制することの構造的限界を指摘し、防衛経済学の生産関数論(レオンチェフ型兵器生産関数)に基づき、先端半導体製造装置(EUV露光等)、軍用赤外線焦点面アレイ(FPA)、高精度慣性計測装置(IMU)、低軌道衛星通信といった「物理的チョークポイント」を基軸とするサプライチェーン抑止戦略の有効性を論証する。さらに、AIを攻撃兵器から受動的・公共的な環境防御インフラへと再定義し、機密計算(Confidential Computing)や暗号学的フェイルセーフ回路(zkVM)を組み込んだ「主権的エッジ(Sovereign Edge)」防護アーキテクチャを提示する。 English Abstract This working paper presents a comprehensive examination of the strategic transformations, international humanitarian law (IHL) challenges, and physical supply chain deterrence mechanisms associated with the rapid technological advancement of Lethal Autonomous Weapon Systems (LAWS) and AI-driven autonomous unmanned aerial vehicles (UAVs) and loitering munitions. In contemporary armed conflicts, robotic warfare has undergone a rapid transition from remotely piloted systems to fully autonomous platforms featuring visual simultaneous localization and mapping (Visual SLAM) in GPS-denied environments, edge-based automatic target recognition (ATR), and decentralized swarm coordination. While this algorithmic acceleration compresses the military decision cycle (OODA loop) to sub-second thresholds, it structurally undermines the doctrine of Meaningful Human Control (MHC) deliberated within the United Nations Convention on Certain Conventional Weapons (CCW) Group of Governmental Experts (GGE), exacerbating digital dehumanization and the accountability gap. Recognizing the fundamental limitations of conventional export control regimes in halting the dissemination of open-source algorithmic weights, this study demonstrates the operational efficacy of a hardware-centric deterrence strategy targeting physical choke-points—including advanced semiconductor lithography, military-grade focal plane arrays (FPAs), and satellite telemetry links—formalized via a Leontief weapon production function. Furthermore, the paper proposes a defensive governance framework that redefines military AI into passive public environmental infrastructure, underpinned by a "Sovereign Edge" architecture integrating confidential computing, zero-knowledge verification (zkVM), and immutable operational circuit breakers.
This publication proposes a definition and a classification of agile software development approaches and analyses ten software development methods that can be characterized as being "agile" against the defined criterion.
P. Abrahamsson, O. Salo, Jussi Ronkainen et al.· arXiv.org· 727 citations· ⚡54
The study shows that agile practices improve both informal and formal communication, but indicates that, in larger development situations involving multiple external stakeholders, a mismatch of adequate communication mechanisms can sometimes even hinder the communication.
M. Pikkarainen, Jukka Haikara, O. Salo et al.· Empirical Software Engineeri...· 401 citations· ⚡48
The results indicate that software engineering work practices are chosen opportunistically, adapted and configured to provide value under the constrains imposed by the startup context.
Nicolò Paternoster, Carmine Giardino, M. Unterkalmsteiner et al.· Information and Software Tec...· 394 citations· ⚡54
The perception of the impact of agile methods is predominantly positive, and several challenge areas were discovered, but based on this study, agile methods are here to stay.
M. Laanti, O. Salo, P. Abrahamsson· Information and Software Tec...· 260 citations· ⚡20
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduOct 2, 2026
Biology doesn't operate in silos, and neither should the AI representation of it. Quine is an early-stage research effort to create a multimodal world model of biology. By connecting insights across biological scales and modalities, Quine helps scientists computationally search a space far larger than intuition allows and prioritize hypotheses before they reach the lab. Experimental results provide important feedback, helping researchers sharpen future research directions. The post Introducing Q…