Structuring The Data Of The Largest Missile Defence Operation In Military History
Between 2024 and 2026, Iran and its proxies launched an unprecedented series of ballistic missile, cruise missile, and drone attacks against Israel and regional targets — the largest sustained missile campaign in modern warfare.
Promise Denied is, at its core, an experiment — an exploration of whether AI can be used to synthesise, aggregate, and visualise a large, fast-moving dataset drawn from multiple conflicting perspectives in near real-time.
The Experiment
The Iran-Israel conflict generates data at a pace that overwhelms traditional OSINT methods. Salvos arrive in rapid succession. Munitions counts vary across sources. Casualty figures are revised. Launch sites are identified days after impact. Israeli, Iranian, American, and international sources offer conflicting accounts of the same events.
This project asks: can a combination of frontier AI models, open-source intelligence, and human editorial oversight produce a structured, geovisualised dataset of a fast-moving conflict — one that is accurate, transparent, and useful?
The answer so far is a qualified yes — and the site you're reading is the result. It is as much an experiment in AI-assisted OSINT methodology as it is a record of the conflict itself.
The Subject
The dataset chosen for this experiment is the largest missile defence operation in military history. Promise Denied geovisualises the devastation caused by Iran's attacks — and the devastation averted thanks to the extraordinary coalition of air defence systems assembled to stop them.
AI-Driven OSINT Synthesis
The data pipeline behind Promise Denied uses a potpourri of AI models and OSINT information sources, combined with human editorial oversight:
Claude Opus 4.6 (Anthropic) for research, data structuring, analysis, and site development. Gemini 3 (Google) for cross-referencing and multi-source verification. Groq for rapid inference on structured data tasks.
IDF, CENTCOM, and IRGC official statements. Reuters, AP, Al Jazeera, Times of Israel. Satellite imagery (Planet Labs, Maxar). CSIS, FPRI, IISS, Alma Center research. Social media verification and munition debris analysis (OSMP).
Multi-model LLM workflows with source grounding — every AI-generated data point is cross-referenced against primary sources before publication. Structured output in JSON, Neo4j, GeoJSON, and KML.
AI handles synthesis and structuring. Humans provide editorial judgement, source verification, and the contextual understanding that models cannot yet replicate — especially in a contested information environment.
Why This Matters
Four rounds of Iranian strikes. 45+ attack salvos. Thousands of munitions. Targets across 12 countries. Arrow, David's Sling, Iron Dome, THAAD, Patriot, Aegis — a layered defence architecture that has intercepted the vast majority of incoming threats.
This is a dataset that deserves to be structured, preserved, and made accessible — both as a historical record and as a demonstration of what open-source intelligence, combined with modern AI tools, can achieve.
