AI Real Estate Glossary

Admin Rot

Low-leverage administrative tasks that drain time from revenue-generating activities

Agentic AI

AI systems that autonomously take actions, make decisions, and complete multi-step tasks

Agentic Operating System

Unified AI system coordinating multiple agents and workflows

AI Agent

Autonomous software powered by LLMs that can execute complex tasks independently

AI Bias

Systematic unfairness in AI outputs due to training data or design

AI Buyer-Seller Matching

Predictive AI that matches buyers to properties using behavioral signals

AI Chatbot

AI-powered conversational interfaces for lead capture and qualification

AI Content Calendar

AI-assisted social media and marketing content scheduling

AI CRM

AI-enhanced customer relationship management for real estate

AI Disclosure & Compliance

Legal requirements for disclosing AI use in real estate

AI Email Marketing

Using AI to create, personalize, and optimize email campaigns

AI-Enhanced Agent

Real estate professional who systematically uses AI for competitive advantage

AI Guardrails

Safety mechanisms that limit what AI can do; prevent harmful/unethical outputs

AI Hallucination

When AI generates false, fabricated, or misleading information confidently

AI Lead Response

Automated, personalized responses to real estate inquiries

AI Listing Automation

Automated listing creation using AI for photos, descriptions, and MLS entry

AI Listing Description

AI-generated property marketing copy optimized for MLS and buyer appeal

AI Market Analysis

AI-assisted comparative market analysis and pricing recommendations

AI Negotiation Assistant

AI-powered tools for negotiation strategy and objection handling

AI Operating System

Unified AI platform that coordinates multiple agents and workflows

AI Personalization

Tailoring communications and recommendations to individual clients with AI

AI Photo Enhancement

AI-powered photo editing and enhancement for real estate listings

AI Property Search

AI-enhanced property search using natural language and predictive matching

AI ROI

Measuring return on investment from AI tool adoption

AI Tenant Screening

AI-powered tenant evaluation with fair housing compliance

AI Transaction Coordination

Automating administrative tasks in real estate transactions

AI Virtual Staging

Using AI to digitally furnish and stage empty property photos

AI Voice Agent

AI systems that handle phone calls for lead capture and qualification

AI Workflow Automation

Designing end-to-end automated processes using AI

AI Writing Detection

Tools and methods to identify AI-generated content

Algorithmic Discrimination

Systematic bias in AI systems affecting housing decisions

Alignment

Ensuring AI systems behave according to human values and intentions

Anthropic API

Developer access to Claude models for building custom applications

API Key

A unique code that authenticates access to AI services

A2A (Agent-to-Agent) Protocol

Standard for communication between autonomous AI agents

Automated Valuation Model (AVM)

Algorithms that estimate property values using data and comparable sales

Autonomous Workflow

AI-driven processes that operate with minimal human intervention

Batch Prompting

Sending multiple related requests to AI simultaneously

Brand Voice

The consistent personality and tone in your communications

Chain of Thought (CoT)

Prompting AI to show reasoning steps ("Let's think step by step")

ChatGPT

OpenAI's conversational AI; excels at logical reasoning, research, and voice mode

Claude

Anthropic's AI assistant; best for writing, coding, and has strongest ethical guardrails

CMA Narrative

AI-generated written analysis for Comparative Market Analysis

Computer Vision

AI that processes and understands visual information from images and video

Context Cards

Structured briefing documents with 4-layer hierarchy for persistent AI context

Context Compaction

Technique for fitting maximum context into fewer tokens

Context Engineering

Pre-loading AI with high-signal tokens—voice, brand DNA, expertise—for 95% usable outputs

Context Rot

Degradation of AI accuracy as token count increases in long conversations

Context Window

Maximum tokens AI can consider in one conversation (GPT-4: 128K, Claude: 200K, Gemini: 2M)

Content Filtering

Systems that block harmful, inappropriate, or policy-violating content

Content Repurposing

Using AI to transform content into multiple formats across platforms

Conversational AI

AI systems that engage in natural dialogue across voice and text channels

Data Privacy

Protecting personal and client information when using AI tools

Deepfake Detection

Tools and techniques to identify AI-generated synthetic media

DeepSeek

Open-source AI model with strong reasoning at low cost

Delegator Mindset

Treating AI as a capable team member you delegate to, not just a tool

Document Automation

AI-powered creation and processing of real estate documents

Drip Campaign

Automated sequences of emails sent over time to nurture leads

Embeddings

Numerical representations of text that capture semantic meaning

Explainability

Ability to understand and explain how AI reaches its conclusions

Few-Shot Prompting

Providing examples in the prompt to guide AI responses

Fine-Tuning

Training a pre-trained model on specific data to specialize its knowledge

The 5 Essentials

Formula for effective prompts: Ask, Audience, Channel, Facts, Constraints

Frequency Penalty

Reduces repetition by penalizing frequently used tokens

Gemini

Google's multimodal AI; best for video/audio analysis and cost-effective enterprise use

Generative Engine Optimization (GEO)

Optimizing content so AI engines recommend and cite you

Google AI Studio

Google's platform for prototyping with Gemini models

Grounding

Connecting AI responses to verifiable source material

HOME Framework

Structured workflow: Hero, Outcome, Materials, Execute

Human-in-the-Loop

AI systems requiring human approval for critical decisions

Inference

The process of an AI generating output from input

Iterative Refinement

Progressively improving outputs through follow-up prompts

Jailbreaking

Attempts to bypass AI safety restrictions through manipulative prompts

Job Audit

Methodology for categorizing tasks by leverage and replaceability

Knowledge Cutoff

The date after which an AI has no training data

Large Language Model (LLM)

AI trained on massive text datasets to understand and generate human-like language

Lead Nurturing

Automated long-term relationship building with leads

Lead Scoring

AI-powered ranking of leads by conversion likelihood

Marketing Automation

AI-driven systems that automate repetitive marketing tasks

Max Tokens

Maximum length of AI response in tokens

MCP (Model Context Protocol)

Standard protocol for connecting AI agents to tools and data sources ("USB-C for AI")

Meta-Prompting

Using AI to iterate on and improve its own prompts

Microsoft Copilot

AI assistant integrated into Microsoft 365 products

Model Parameters

The learned values (weights) that define an LLM's behavior (GPT-4: ~1.76T)

MLS AI Integration

AI features built into MLS platforms for listing automation

Multi-Agent Systems

Multiple AI agents working together on shared objectives

Multimodal AI

AI that processes multiple input types: text, images, audio, video

Natural Language Processing (NLP)

AI field enabling computers to understand human language

Negative Prompting

Specifying what NOT to include in outputs

Neural Network

Computing system inspired by biological neurons; foundation of modern AI

Newsletter Segmentation

Using AI to segment email lists by behavior and demographics

AI Objection Handling

Using AI to prepare responses to common buyer and seller objections

AI Open House Strategy

Using AI for open house preparation, promotion, and follow-up

One-Shot Prompting

Providing exactly one example to guide AI

OODA Loop for AI

Military decision-making adapted for AI workflow: Observe, Orient, Decide, Act

OpenAI API

Developer access to GPT models for building custom applications

Out-of-Band Verification

Confirming requests through separate communication channels

Perplexity

AI-powered search engine combining conversational AI with real-time web search

Pre-training

Initial training of LLMs on massive datasets before specialization

Predictive Analytics

Using data and ML to forecast market trends and seller likelihood

Presence Penalty

Encourages topic diversity by penalizing already-mentioned concepts

Prompt

The input text given to an AI to generate a response

Prompt Caching

Storing and reusing prompt parts to reduce cost and latency

Prompt Chaining

Breaking complex tasks into sequential prompts

Prompt Engineering

The practice of crafting effective AI prompts for desired outputs

Prompt Injection

Attack technique where malicious instructions are hidden in prompts

Prompt Library

Curated collection of tested, reusable AI prompts

RAG (Retrieval-Augmented Generation)

Combining LLMs with external knowledge retrieval for accurate, sourced responses

Responsible AI

Framework for developing AI that is ethical, fair, and accountable

RLHF

Reinforcement Learning from Human Feedback—training using human preferences

Role Prompting

Assigning AI a specific persona or expertise ("You are a luxury real estate agent...")

Self-Consistency

Running multiple AI generations and selecting the most consistent answer

Semantic Search

Search based on meaning rather than exact keyword matching

Sentiment Analysis

AI technique for detecting emotions and attitudes in text

Strategic Displacement

Ruthlessly replacing low-leverage tasks with AI to focus on revenue-generating activities

Style Match

Technique for extracting voice from writing samples

Synthetic Media Disclosure

Legal requirements for disclosing AI-generated or altered media

System Prompt

Hidden instructions that shape AI behavior before user interaction

Temperature

Controls AI "creativity" and randomness (0-2 scale; lower = more predictable)

Text-to-Image AI

AI that generates images from text descriptions

The Gap

The widening divide between AI-Enhanced and Traditional agents (68% use, 17% see results)

The 80/20 Rule for AI

AI automates the 80% of tasks so you can focus on the 20% that generates 80% of results

Tiny Stack

Minimalist AI tool philosophy—master few vs. subscribe to many

Token

Unit of text AI processes (~0.75 words); determines pricing and context limits

Tool Use / Function Calling

AI capability to invoke external tools, APIs, and actions

Top-P (Nucleus Sampling)

Limits token selection to top cumulative probability threshold

Top-K

Limits token selection to K most likely options

Traditional Agent

Real estate professional using conventional methods without AI systems

Transformer

The neural network architecture underlying modern LLMs

Tree of Thought

Generating branching paths of reasoning for complex problems

Vector Database

Database optimized for storing and searching embeddings

AI Video Script Generation

Using AI to create scripts for property tours and market updates

Zero-Shot Prompting

Asking AI to complete tasks without providing examples