THE BLACK BOX PROBLEM IN AI AND ITS IMPACT ON ATTRIBUTION IN CYBER CONFLICTS: INTERNATIONAL LAW’S RESPONSE TO UNTRACEABLE DECISION-MAKING

AUTHOR – ALLEN BENNY, BBA LLB, SCHOOL OF LAW, CHRIST (DEEMED TO BE UNIVERSITY), BENGALURU

BEST CITATION – ALLEN BENNY, THE BLACK BOX PROBLEM IN AI AND ITS IMPACT ON ATTRIBUTION IN CYBER CONFLICTS: INTERNATIONAL LAW’S RESPONSE TO UNTRACEABLE DECISION-MAKING, INDIAN JOURNAL OF LEGAL REVIEW (IJLR), 6 (2) OF 2026, PG. 664-671, APIS – 3920 – 0001 & ISSN – 2583-2344

Abstract

The rapid militarization of artificial intelligence in cyber operations highlights a growing concern in international law. The “black box” nature of deep-learning systems means that their autonomous decision-making processes are untraceable, even to their creators. This type of technological opacity undermines attribution, a doctrine that is concerned with proof of intent, control (effective, overall, or in a causal sense), and the law of international responsibility, including the ARSIWA, the Tallinn Manual 2.0, and key case law, including the Nicaragua and Tadić cases. When fully autonomous AI systems behave erratically and unpredictably, the law of international responsibility and international humanitarian law (IHL) are left with critical gaps, flouting the principles of sovereignty, distinction, proportionality, and the law of armed conflict (meaningful human control).

Employing a doctrinal, comparative, and interdisciplinary methodology, this paper analyses how algorithmic opacity disrupts legal attribution in cyber conflicts and evaluates whether Explainable AI (XAI) techniques can restore transparency and traceability. While XAI cannot eliminate the black box entirely, it offers practical audit trails and evidentiary support for Article 36 weapons reviews and post-incident investigations. The findings demonstrate that current international legal instruments remain ill-equipped for non-human agency. The paper concludes that meaningful reform—incorporating a legal duty of explainability, constructive control standards, and institutional oversight mechanisms—is essential to preserve accountability in the age of autonomous cyber warfare.

Keywords: Artificial Intelligence, Black Box Problem, State Responsibility, Cyber Warfare, Attribution, Explainable AI, Constructive Control.