Artificial Intelligence
How AI and autonomous systems work, where they are being fielded, and what they mean for military and enterprise decision-makers. Courses cover foundations, applications, and the policy and acquisition questions that follow.
AI Agents, Automation, and Workflow Integration
Covers how AI agents and automation tools plan tasks, call tools, and integrate with existing workflows and systems. Learners assess where agents add value, how to scope their permissions, and how to keep humans accountable for outcomes.
AI Assurance and Adversarial ML - Attacking and Defending AI Systems
Covers the main classes of attack on machine learning systems, including evasion, poisoning, model extraction, and prompt injection. Learners learn defenses and assurance methods that establish justified confidence in AI system behavior.
AI Governance, Policy, and Responsible AI
Covers the policies, frameworks, and oversight structures that govern AI use in government and industry, including the NIST AI Risk Management Framework. Learners learn to translate responsible AI principles into concrete development and deployment practices.
AI Incident Response and Model Compromise
Covers how to detect, contain, and recover from AI incidents, including model compromise, data poisoning, and harmful outputs. Learners build response plans that add AI-specific steps to existing incident response processes.
AI Remediation - Diagnosing and Fixing Underperforming Models
Covers structured methods for finding why a model or AI system underperforms, from data quality and distribution shift to integration errors. Learners practice isolating root causes and selecting a fix matched to each.
AI Test & Evaluation Beyond Benchmarks
Examines the limits of standard benchmarks and methods for evaluating AI under realistic operational conditions. Learners design evaluations that measure mission-relevant performance, robustness, and failure modes.
Compute, Chips, and the AI Infrastructure Supply Chain
Covers the hardware and infrastructure behind AI, including semiconductors, data centers, power, and the global supply chain that produces them. Learners examine chokepoints, export controls, and the national security implications of compute access.
Data Infrastructure for Operational AI
Covers the pipelines, storage, access controls, and architectures needed to deliver data to AI systems in operational settings. Learners work through challenges such as multiple classification levels, disconnected environments, and data interoperability.
Data Labeling, Curation, and Provenance
Covers how training and evaluation data are collected, labeled, cleaned, and documented. Learners learn to track data provenance and quality so model behavior can be traced and trusted.
Edge AI in Constrained and Contested Environments
Covers deploying AI on devices with limited compute, power, and connectivity, including techniques such as model compression. Learners examine how edge systems operate under degraded communications and adversary interference.
From Pilot to Program of Record - Transitioning AI Into Production
Examines why AI pilots often fail to transition and the funding, acquisition, and organizational steps that support production. Learners plan transitions that account for sustainment, authorization, and user adoption.
Generative AI and Large Language Models for Mission Use
Explains how large language models and other generative AI systems work and where they fit mission tasks. Learners assess risks such as hallucination, data leakage, and prompt injection and learn safe deployment practices.
Human-Machine Teaming and Decision Support
Covers how people and AI systems share tasks and decisions, including trust calibration, interface design, and automation bias. Learners design decision support that improves human judgment under time pressure.
Introduction to Data Science and AI
Introduces core concepts in data science, statistics, and machine learning for non-specialists. Learners gain the vocabulary and judgment needed to work with technical teams and assess AI proposals.
Machine Learning for Mission Problems
Covers how to frame operational problems as machine learning tasks and select appropriate methods and data. Learners work through the steps from problem definition to a validated model.
MLOps in High-Consequence Settings - Monitoring, Drift, Retraining, and Retirement
Covers the practices for deploying, monitoring, and maintaining models in regulated and high-consequence environments. Learners learn to detect drift, manage retraining and rollback, decide when to retire a model, and document each decision for oversight.