Anticipatory Intelligence

Anticipatory intelligence focuses on characterizing and reducing uncertainty by providing decision makers with timely and accurate forecasts of significant global events. This research explores or demonstrates the feasibility of revolutionary concepts that may deliver real-time indications and warning, in context, to support rapid, nuanced understanding by intelligence consumers.

Our programs are technically diverse, but each program:

  • Develops technologies to generate timely forecasts for well-defined events and their characteristics (e.g., who, what, when, where, and how).
  • Uses a rigorous, open and ongoing test and evaluation process.
  • Has metrics that include lead time, accuracy, false positive and false negative rates, and are calculated by comparing forecasts to real-world events.
  • Communicates forecasts in context.

Key research areas include forecasting events related to science and technology (S&T); social, political, and economic crises; epidemiology and biosecurity; counterintelligence; and cybersecurity.

Current Research | View Past Research

ProgramResearch AreaProgram Manager
CAUSE Cybersecurity, cyber-event forecasting, cyber-actor behavior and cultural understanding, threat intelligence, threat modeling, cyber-event coding, cyber-kinetic event detection Robert Rahmer
CREATE Forecasting, logic and critical thinking, human judgment Steven Rieber
FUSE Technical emergence, text analytics, knowledge discovery, big data, social network analysis, natural language processing, forecasting, machine learning Robert Rahmer
HFC Forecasting, human judgment, machine learning, decision making, human/machine interfaces, text analysis Seth Goldstein
Mercury SIGINT analytics, event forecasting, machine learning, streaming data, data fusion, weapons of mass destruction, chemical/biological warfare, human biomarkers, emerging biotechnologies Kristen Jordan
SCITE Engineering enterprises that detect low probability events with low accuracy sensors, innovative research methods to evaluate analytic and forecasting tradecraft, innovative statistical methods to estimate performance of systems addressing complex analysis and forecasting problems, scientific research on organizational lessons-learned methods, evidence-based forecasting methods, inductive logic, probabilistic reasoning and its application to analytic tradecraft Paul Lehner

Past Research

ProgramResearch Area
ACE Forecasting, human judgment, machine learning, logic, critical thinking
ForeST Forecasting, human judgment, machine learning, technical emergence, text analytics, big data, natural language processing
OSI Large/errorful data sets, forecasting, public health, machine learning