Insurance AI Dictionary
Reference Guides
Executive reference publications on AI for Property & Casualty insurance—from foundations through governance.
Looking for implementation centers or practical tools?
Use the Navigator to find the right package by role, objective, and stage of AI adoption.
All reference guides
35 publications
Artificial Intelligence
A Practical Guide for Property & Casualty Insurance
Artificial intelligence is not a single product or model. It is a family of technologies that …
Machine Learning
The Predictive Engine Powering Modern Insurance AI
Machine learning enables insurers to learn patterns from historical data and apply those …
Generative AI
Creating New Content While Transforming Insurance …
Generative AI is a class of artificial intelligence that creates new content - including …
Large Language Models
How Language-Based AI Understands and Supports …
A large language model is an AI model trained on very large collections of text so it can …
Natural Language Processing
Turning Insurance Language Into Actionable …
Natural language processing converts human language into structured, searchable, and …
Retrieval-Augmented Generation
Grounding Generative AI in Trusted Insurance …
Retrieval-Augmented Generation is an architecture that retrieves relevant information from …
Predictive AI
Turning Insurance Data into Earlier, Explainable …
Predictive AI uses historical and current information to estimate likely future outcomes so …
Deep Learning
Advanced Neural Networks for Complex Insurance …
Deep learning is a specialized form of machine learning that uses multi-layer neural networks …
Neural Networks
How Layered Learning Models Recognize Patterns and …
A neural network is a machine learning model composed of interconnected computational units …
AI Agents
Autonomous and Assistive AI for Governed Insurance …
An AI agent is a software system that can interpret a goal, plan and execute permitted …
Supervised Learning
Learning from Known Outcomes to Improve Insurance …
Supervised learning is a machine learning approach in which a model learns from historical …
Unsupervised Learning
Discovering Hidden Patterns, Segments, and …
Unsupervised learning analyzes data without predefined outcome labels to reveal natural …
Reinforcement Learning
Learning Sequential Decisions Through Feedback, …
Reinforcement learning is a method for improving sequential decisions by learning from …
Computer Vision
Turning Insurance Images and Video into Actionable …
Property & Casualty insurance is inherently visual. Damage photographs, vehicle images, …
Optical Character Recognition
Turning Scanned Insurance Documents into …
Optical Character Recognition is the technology used to detect and convert visible characters …
Intelligent Document Processing
From Documents to Structured, Evidence-Linked …
Intelligent Document Processing is an end-to-end capability that receives documents, …
Embeddings
A Practical Guide for Property & Casualty Insurance
Embeddings are numerical representations that place words, passages, documents, images, or …
Vector Databases
A Practical Guide for Property & Casualty Insurance
A vector database is a specialized data platform designed to store high-dimensional …
Knowledge Graphs
A Practical Guide for Property & Casualty Insurance
Knowledge graphs provide a structured way to represent real-world entities, their attributes, …
Semantic Search
A Practical Guide for Property & Casualty Insurance
Semantic search retrieves information based on meaning, context, and conceptual …
Fine-Tuning
A Practical Guide for Property & Casualty Insurance
Fine-tuning is the controlled process of further training a pre-trained model on curated …
Prompt Engineering
A Practical Guide for Property & Casualty Insurance
Can an insurer convert general-purpose AI capability into repeatable, controlled assistance …
Model Context Protocol
Connecting Governed AI Assistants and Agents to …
Model Context Protocol (MCP) is an open protocol that standardizes how AI applications …
Explainable AI
A Practical Guide for Property & Casualty Insurance
Explainable AI (XAI) is the discipline of making AI outputs sufficiently understandable for …
Responsible AI
A Practical Guide for Property & Casualty Insurance
Responsible AI is the disciplined design, deployment, and oversight of artificial …
AI Governance
Strategy • Accountability • Risk-Tiered Controls • …
AI governance is the enterprise system of policies, decision rights, roles, controls, …
AI Guardrails
A Practical Guide for Property & Casualty Insurance
AI guardrails are the policies, technical controls, workflow constraints, monitoring …
AI Hallucinations
A Practical Guide for Property & Casualty Insurance
An AI hallucination is an output that sounds plausible but is unsupported, fabricated, …
AI Bias
A Practical Guide for Property & Casualty Insurance
AI bias is a systematic tendency in data, models, or workflows that can produce distorted, …
AI Model Drift
A Practical Guide for Property & Casualty Insurance
AI model drift is a material change in the relationship between an AI system, its inputs, its …
Data Quality
A Practical Guide for Property & Casualty Insurance
Data quality is the fitness of data for a defined business purpose. It is measured against …
Feature Engineering
A Practical Guide for Property & Casualty Insurance
Feature engineering is the disciplined process of converting raw insurance data into …
Synthetic Data
A Practical Guide for Property & Casualty Insurance
Synthetic data is artificially generated information designed to reflect the statistical …
AI Agents for Insurance
How Autonomous and Human-Governed AI Agents Can …
AI agents represent a shift from isolated prediction or content generation toward governed …
Agentic AI
The Executive Guide to Autonomous AI Systems in …
Agentic AI refers to artificial intelligence systems that can autonomously pursue goals, plan …