How BattlePolicy uses AI, sources its facts, and stays accountable to readers.
BattlePolicy both synthesizes and reports. We read widely across outlets and primary documents, and we report on and analyze the development of the defense industry — the companies, capabilities, funding, and procurement behind modern war — and connect it to what is happening on the front. Factual claims are attributed to a named outlet or primary source, inline in the prose.
We use AI tools to help us research and surface sources, draft, and synthesize across a large volume of reporting — it lets a small team move fast and cover more ground. The editorial judgment is human: what is worth covering, what it means, and what gets published. We disclose when analysis is AI-assisted, and no factual claim is published on an AI's say-so alone.
Every substantive claim is checked by a human editor against primary documents, official statements, or independent reporting — not press releases on their own. This is a contested information space, so high-stakes claims (strike results, casualty and loss figures, capability "firsts") need multiple independent sources, or they are explicitly hedged and single-attributed so you can weigh them yourself.
Every piece carries a human byline and is clearly labeled as news or analysis. We attribute relentlessly — a named source on every hard claim — and we never present synthesis as first-hand reporting.
If we get something wrong, we fix it promptly and note the change. Spot an error, or have a question about how we work? Email editor@battlepolicy.com.