Autonomous Warfare: Competition Trumps Treaties
During Operation EPIC FURY in the Iran War (February–March 2026), the U.S. and Israel deployed the Maven Smart System, which integrated Anthropic's Claude AI model to accelerate target processing from under 100 per day to up to 1,000 per day, compressing the kill chain from days to seconds.
US, China, Russia Resist Binding AWS Rules
The Arms Control Association and Lawfare report that over 120 countries, alongside the International Committee of the Red Cross (ICRC) and various NGOs, argue existing International Humanitarian Law (IHL) is insufficient for autonomous weapon systems (AWS); this stance contrasts with the United States, China, and Russia, who resist new binding international restrictions and prefer voluntary national guidelines over multilateral constraints, asserting that existing IHL is sufficient if AWS comply with established principles. These same 120-plus countries, the ICRC, and various NGOs further contend that IHL is insufficient for AWS due to algorithmic unpredictability and accountability gaps, pushing for strict "meaningful human control" (MHC) thresholds to prohibit fully autonomous lethal systems. Captain K.S. Vikramaditya pointed out that Germany's coalition agreement, for example, rejects lethal autonomous systems without human control, whereas U.S. policy permits them with "appropriate levels of human judgment."
Major Powers Converge on Machine-Speed Interoperability
A University of Copenhagen analysis demonstrated that autonomous systems offer a decisive advantage by enabling "machine-speed decision-making" that outpaces human cognition and traditional command chains; this capability, Genesys Defense reported, compels major powers to converge on standardized operational standards based on shared technical behaviors and machine-speed interoperability for military superiority. The Lieber Institute for Law & Land Warfare explains that operational standards in autonomous warfare are practically defined by interoperable technical performance and algorithmic decision-making speeds, rather than by unified legal thresholds of meaningful human control. Autonomous systems also provide a decisive advantage by enabling "machine-speed decision-making" that outpaces human cognition and traditional command chains, a University of Copenhagen analysis revealed. Genesys Defense and a peer-reviewed article indicate that operational predictability emerges from these shared technical behaviors and algorithmic efficiency, rather than from harmonized diplomatic agreements.
US, Russia, China Converge on Multi-Domain Attacks
The Bulletin of the Atomic Scientists reports that contemporary strategic competition has driven a broader convergence in operational concepts among the U.S., Russia, and China; all three nations now view future conflict as a system confrontation and emphasize deep multi-domain attacks involving kinetic, electronic, and cyber capabilities. The U.S. Department of State and Gettysburg College documented how the Cold War nuclear arms race, marked by the symmetrical adoption of atomic and hydrogen bombs by the United States and the Soviet Union, led to the shared strategic doctrine of Mutually Assured Destruction (MAD). The University of Notre Dame explains that this historical precedent demonstrates how rival powers mirror each other's capabilities and converge on shared doctrinal understandings. The University of Notre Dame also observed that NATO's 1999 intervention in Serbia prompted a major revision of Russia’s National Security Concept and Military Doctrine in April 2000, illustrating how a rival alliance's operational actions can force doctrinal convergence.
IDF Lavender AI Raised IHL Concerns
The Verge highlighted that the Israel Defense Forces (IDF) deployed the Lavender AI targeting program in the Israel-Gaza Conflict (since October 2023) to rapidly identify tens of thousands of individuals, a move that raised significant concerns regarding IHL principles of distinction and proportionality. UNIDIR documented a U.S. Tomahawk missile strike on an Iranian elementary school, which occurred due to outdated intelligence processed at machine speed before human verification could catch it, thereby exposing the lack of standardized testing and certification for AI targeting. In the Russia-Ukraine Conflict (since 2022), The Verge reported on Ukraine's Delta system, which fused multi-source intelligence to reduce the kill chain from 72 hours to 2 minutes. Captain K.S. Vikramaditya observed that while NATO achieved "engineering interoperability" in multinational operations, it struggled with "authority interoperability" due to differing national thresholds for human control.
Major Powers Adopt Machine-Speed Decision Cycles
The Lieber Institute for Law & Land Warfare and the U.S. Naval War College explain that AI integration shifts humans from active evaluators to procedural approvers, thereby increasing the risk of unintended escalation due to algorithmic brittleness and expanded cyberattack surfaces; this development compels major powers to adopt interoperable, machine-speed decision cycles to maintain parity or superiority, compressing diplomatic warning times and enabling machine-to-machine decision cycles that often bypass deliberate human intervention. Furthermore, these institutions also point out that this integration of AI shifts humans from active evaluators to procedural approvers, increasing the risk of unintended escalation due to algorithmic brittleness and expanded cyberattack surfaces. Consequently, mutual predictability in autonomous conflict will rely on shared technical interoperability and algorithmic optimization driven by strategic competition, rather than on harmonized legal frameworks.
Comments ()