# AI and the Future of Terrorism Terrorism historically exploited informational asymmetry. Small networks could hide inside weak ties, aliases, fragmented records, jurisdictional seams, data volume, and limited human attention. AI reduces those advantages by making dynamic graphs, uncertain identity, multimodal fusion, anomaly detection, latent-state estimation, conditional forecasting, and continuous updating computationally tractable at larger scales. The thesis is not that AI makes terrorism mathematically impossible. It is that AI changes the concealment economy. Operational security becomes more expensive when relationships can be resolved across modalities, temporal patterns can be compared, alternative identities can be linked, and sparse observations can update network hypotheses in near real time. Read [[wiki/AI and the End of Terrorism|AI and the End of Terrorism]], [[wiki/Counterterrorism Predictive Graph|Counterterrorism Predictive Graph]], [[wiki/Predictive Intelligence Loop|Predictive Intelligence Loop]], [[wiki/Graph Neural Network|Graph Neural Network]], [[wiki/Entity Resolution|Entity Resolution]], [[wiki/Multimodal Fusion|Multimodal Fusion]], [[wiki/Link Prediction|Link Prediction]], and [[wiki/Uncertainty-Aware Counterterrorism Analysis|Uncertainty-Aware Counterterrorism Analysis]]. The system remains bounded by the difference among observation, inference, prediction, and authorized intervention. AI can compress evidence and expose structure; it cannot supply lawful authority or abolish uncertainty. ## Relationships [[collections/Terrorism, Counterterrorism, and the Intelligence Environment|Collection landing page]] · [[wiki/Counterterrorism Collection - Artificial Intelligence|Artificial Intelligence]] · [[wiki/Counterterrorism Collection - Compliance and Trust Infrastructure|Compliance and Trust Infrastructure]]