AI and Collision Repair Glossary

A

ADAS

Advanced driver assistance systems are vehicle technologies that help a driver avoid collisions or perform certain driving tasks using cameras, radar, lidar, ultrasonic sensors, software, and other components.

ADAS Calibration

ADAS calibration is the process of testing and precisely aligning a vehicle's cameras, radar units, and other driver-assistance sensors so they interpret their surroundings correctly.

AI Assistant

An AI assistant is software that uses artificial intelligence to answer questions, organize information, draft content, or help a person complete a task through conversational or other interfaces.

AI Estimating

AI estimating uses artificial intelligence to analyze claim information, vehicle images, repair data, or historical patterns and assist with creating or reviewing a collision repair estimate.

AI-OCR

AI-OCR is optical character recognition enhanced by artificial intelligence. It reads text from photos, scans, and PDFs, then identifies and organizes useful information instead of merely copying the characters.

Algorithm

An algorithm is a defined set of rules or calculations that a computer follows to process information or solve a problem.

Application Programming Interface

An application programming interface is a defined way for separate software systems to exchange data or request functions from one another.

Artificial Intelligence

Artificial intelligence is a broad category of computer technology designed to perform tasks that normally require human perception, reasoning, language understanding, prediction, or decision support.

Automated Damage Appraisal

Automated damage appraisal is the use of software to assess vehicle damage and produce or support an initial evaluation of repair needs, severity, or cost.

Automation

Automation is the use of technology to complete a task or workflow with reduced manual effort. Automation may use fixed rules, artificial intelligence, or a combination of both.

B

Bias

AI bias is a systematic tendency for a model's results to favor, disadvantage, or misrepresent certain situations because of its training data, design choices, or how it is used.

C

Claim Risk Detection System

A claim risk detection system is software that reviews claim information and flags cases that may need closer human review. It can use business rules, statistical models, or AI to identify missing information, inconsistencies, unusual patterns, or other risk indicators.

Claims Triage

Claims triage is the process of sorting new claims by factors such as severity, complexity, repairability, urgency, and the type of review they require.

Cloud Computing

Cloud computing delivers software, data storage, or computing power over the internet instead of relying entirely on a computer or server located at the user's facility.

Computer Vision

Computer vision is a field of artificial intelligence that enables software to analyze and interpret photographs, video, and other visual information.

Confidence Score

A confidence score is a number that represents how certain an AI system is about a particular prediction or classification. It is not a guarantee that the result is correct.

D

Damage Detection

Damage detection is the use of image analysis or other sensors to locate and classify visible vehicle damage such as dents, scratches, cracks, and broken components.

Data Privacy

Data privacy concerns how personal, vehicle, claim, image, and business information is collected, used, shared, retained, and protected.

Deep Learning

Deep learning is a type of machine learning that uses neural networks with many processing layers to learn complex patterns from large amounts of data.

Diagnostic Scan

A diagnostic scan uses a scan tool to communicate with a vehicle's electronic control modules and retrieve information such as diagnostic trouble codes, system status, and identification data.

Digital Twin

A digital twin is a digital representation of a physical object or process that can be updated with data to reflect its condition, configuration, or behavior.

E

Explainable AI

Explainable AI refers to methods that help people understand why an artificial intelligence system produced a particular result or recommendation.

F

First Notice of Loss

First notice of loss is the initial report to an insurer that an accident, damage event, or other covered loss has occurred.

Fraud Detection

Fraud detection uses rules, data analysis, and sometimes artificial intelligence to identify claims, images, documents, or behavior that may require additional investigation.

G

Generative AI

Generative AI is artificial intelligence that creates new content—such as text, images, audio, video, or code—based on patterns learned from training data and instructions supplied by a user.

H

Hallucination

An AI hallucination is a response that sounds plausible but contains invented, unsupported, or incorrect information.

Human-in-the-Loop

Human-in-the-loop describes a process in which a person reviews, corrects, approves, or otherwise participates in decisions made or supported by an automated system.

I

Image Classification

Image classification is a computer-vision task that assigns an entire image to one or more categories.

Image Segmentation

Image segmentation is a computer-vision technique that assigns image pixels to specific objects or regions so their precise shapes and boundaries can be analyzed.

Inference

Inference is the stage when a trained AI model processes new information and produces a prediction, classification, recommendation, or generated response.

L

Large Language Model

A large language model is an AI model trained on extensive text data to understand and generate language, answer questions, summarize information, and perform other language-based tasks.

M

Machine Learning

Machine learning is a type of artificial intelligence in which a computer model learns patterns from examples or data instead of relying only on rules written explicitly by a programmer.

Model

An AI model is a mathematical system trained or configured to recognize patterns and produce outputs such as classifications, predictions, recommendations, or generated content.

N

Natural Language Processing

Natural language processing is a field of artificial intelligence focused on enabling computers to analyze, understand, and generate human language.

Neural Network

A neural network is a machine-learning model made of connected processing units arranged in layers that learn patterns by adjusting the strength of those connections during training.

O

Object Detection

Object detection is a computer-vision task that identifies objects in an image and indicates where each object is located, commonly with a box or other boundary.

OEM Repair Procedures

OEM repair procedures are instructions and specifications published by a vehicle's original equipment manufacturer for inspecting, repairing, replacing, measuring, calibrating, and validating vehicle systems and structures.

Optical Character Recognition

Optical character recognition converts text visible in an image or scanned document into machine-readable characters that software can search or process.

P

Prompt

A prompt is the instruction, question, context, or example supplied to a generative AI system to guide the response it produces.

R

Retrieval-Augmented Generation

Retrieval-augmented generation is a method in which an AI system retrieves information from selected documents or databases and uses that material as context when generating an answer.

S

Straight-Through Processing

Straight-through processing is a workflow in which qualifying transactions or claims move through defined steps automatically with little or no manual handling.