# Adversarial Intent **Entity class:** Inferred human state **Adversarial intent** is an analytic hypothesis about an actor's purpose, planned direction, or willingness to produce harm. It is not directly observable in the same way as an encounter, transaction, or sensor reading. DTRA's Cognitive and Information Science material treats improved understanding of WMD-related human motivation and decision factors as a research problem within physical and social networks. Intent therefore belongs to the inferred-state layer. Evidence can increase or decrease the plausibility of competing intent hypotheses, but graph proximity, ideological exposure, or anomalous behavior alone does not convert intent into an observed fact. ## Relationships - **research context:** [[wiki/Cognitive and Information Science|Cognitive and Information Science]] and [[wiki/Defense Threat Reduction Agency|Defense Threat Reduction Agency]]. - **estimated through:** [[wiki/Inference Under Partial Observation|Inference Under Partial Observation]], [[wiki/Competing Hypotheses|Competing Hypotheses]], and [[wiki/Behavioral Model|Behavioral Model]]. - **bounded by:** [[wiki/Prediction versus Determination|Prediction versus Determination]]. ## Sources - [Basic Research for Countering Weapons of Mass Destruction — DTRA](https://www.govinfo.gov/content/pkg/GOVPUB-D15-PURL-gpo18268/pdf/GOVPUB-D15-PURL-gpo18268.pdf) **As of:** 2026-09-23