Leveraging AI for Strategic Cyber Diplomacy and International Norms Compliance

Thesis Abstract

This study investigates how artificial intelligence (AI) can enhance strategic cyber diplomacy and the enforcement of international norms in cyberspace amidst rising state and non-state cyber activities and evolving normative frameworks. The problem addressed is the gap between rapidly advancing cyber capabilities and the slow, fragmented application of international norms and diplomatic instruments, which undermines deterrence, accountability, and collective security. The aim is to develop an AI-enabled framework that supports strategic cyber diplomacy by (i) discovering and assessing normativity attributions across actors, (ii) automating monitoring and verification of norm compliance, and (iii) informing negotiation stances and coalitional coordination. Specific objectives are to (a) map existing international norms and their normative tensions using AI-assisted text mining; (b) evaluate the effectiveness of AI-driven indicators for cyber norm compliance through a cross-national dataset of 60 states over the period 2015–2025; (c) design a decision-support prototype that sequences diplomatic communications, public attribution, and coalition formation; (d) assess the ethical and governance implications of AI-assisted diplomacy; and (e) propose policy-ready recommendations for regional and global institutions. The methodology adopts a mixed-methods design combining quantitative and qualitative strands. The population comprises state actors, international organizations, and major cyber-enabled non-state groups engaged in cyber diplomacy. A stratified purposive sample of 60 states is selected to reflect diverse governance models, capabilities, and normative alignments, with 30 high-capability and 30 middle/low-capability states. Data collection employs (i) a structured content corpus of 2,400 official statements, UN resolutions, and bilateral communications from 2015–2025; (ii) cyber incidents and attribution datasets from the Global Cyber Security Dataset and the ITU; and (iii) interviews with 40 senior policymakers, diplomats, and technology advisors. Instruments include a coder-validated norm-aptitude coding scheme for content analysis, AI-enabled text-mining pipelines for semantic network mapping, and a structured interview guide. Validity and reliability are ensured through triangulation, inter-coder reliability checks (Cohen’s kappa > 0.75), and pilot testing of the AI sentiment and norm-detection models. Data analysis proceeds in three layers. First, content analysis and topic modeling (LDA) identify predominant normative themes, tensions, and framing. Second, AI-assisted indicator construction uses supervised machine learning (random forest and gradient boosting) to predict norm compliance scores, validated against expert assessments. Third, the study develops and tests an agent-based model of cyber diplomacy scenarios to evaluate strategic outcomes of different negotiation tactics and coalition configurations, with results analyzed via regression analysis and scenario testing. The analytical framework integrates construct validity from the liberal institutionalism and constructivist theory lenses, drawing on the hypothetical relevance of (i) deterrence theory in cyber diplomacy and (ii) the social constructivist emphasis on norms and identities, complemented by the cultural theory of risk governance to interpret cross-national differences. A conceptual model illustrates the links among AI-enabled norm discovery, monitoring, decision support, and diplomatic outcomes. Key findings are expected to show that AI-assisted norm discovery improves the identification of tacit norms and convergence zones across actors, that automated monitoring enhances timely attribution and reduces escalation, and that AI-informed negotiation support increases coalition coherence and norm compliance rates by 15–25 percentage points in simulated diplomatic exchanges. The study anticipates variability across governance regimes, with democracies exhibiting faster norm adaptation and more transparent experimentation with AI-enabled diplomacy, while autocratic regimes retain strategic ambiguity in attribution, creating a nuanced landscape for alliance-building and deterrence. The contribution to knowledge lies in (i) operationalizing AI within the practice of cyber diplomacy and international norm compliance, (ii) providing a replicable methodological blueprint for integrating AI into diplomatic decision-support, (iii) advancing theoretical understanding of how AI affects norm diffusion, attribution, and coalition dynamics in cyberspace, and (iv) offering evidence-based policy recommendations for international institutions, regional organizations, and state actors. The main conclusion posits that AI can significantly enhance strategic cyber diplomacy when embedded within transparent governance frameworks, with carefully designed human-in-the-loop processes ensuring accountability. Recommendations include developing standardized AI ethics guidelines for diplomacy, investing in interoperable norm-monitoring platforms within UN and regional bodies, and fostering multilateral pilots to test AI-enabled negotiation protocols under independent oversight.