Research Reports
Long-form research publications with comprehensive analysis, methodology, and findings.
AI-Mediated Property Discovery Report 2026
Evidence from 50 Markets, Thousands of AI Responses, and Observed Property Selection Behavior
The AI-Mediated Property Discovery Report 2026 presents the first comprehensive observational study of how AI systems discover, evaluate, compare, and select properties across diverse markets. Through systematic observation of AI response patterns across 50 real estate markets, thousands of AI responses, and documented selection events, this report establishes the empirical foundation for understanding AI-mediated property discovery. The report analyzes property selection behavior, identifies top selection signals, examines explainability patterns, measures representation effects, and documents citation sources that inform AI decision-making.
AI Summary
AI-mediated property discovery consistently relies on structured, explicit, and context-rich information when evaluating and selecting properties across observed markets.
Representation Gap Report 2026
Measuring the AI Discoverability of Modern Property Records
The Representation Gap Report 2026 examines the disconnect between traditional property listing practices and AI-mediated discovery requirements. Through analysis of 50 real estate markets and systematic observation of AI response patterns, we identify specific representation weaknesses that prevent properties from being selected by AI systems. The report establishes the Representation Efficiency Score (RES) as a standardized measure of how effectively a property record enables AI selection.
AI Summary
Properties with complete, structured VPR representation appear in AI selections 3.2x more frequently than comparable properties without.
VPR Selection Experiment 2026
Experimental Study of Representation Structure Effects on AI Property Selection
The VPR Selection Experiment 2026 evaluates the effect of property representation structure on AI-mediated property selection. Equivalent properties were represented using both traditional listing formats and Verified Property Records (VPRs) and evaluated across standardized AI selection environments. This controlled experimental design isolates representation structure as the independent variable while holding property attributes, selection scenarios, and AI systems constant.
AI Summary
Structured VPR representations were selected 3.24x more frequently and explained with 68.3% higher specificity than equivalent traditional listings across controlled experimental conditions.
AI Selection Signals Report 2026
Measured Analysis of Property Attributes Driving AI-Mediated Selection
The AI Selection Signals Report 2026 identifies and ranks the property attributes that most strongly influence AI-mediated property selection behavior. Through systematic measurement of AI response patterns across 50 markets and standardized analysis of surfaced properties, we establish which attributes serve as primary selection signals across hospitality and real estate verticals.
AI Summary
AI systems consistently prioritize explicit, structured, verifiable property attributes over implicit or narrative descriptions, with location context being the strongest selection signal across both hospitality and real estate verticals.
Machine Readability Validation Study 2026
Validation of the MRI Framework Against Observed AI Selection Outcomes
The Machine Readability Validation Study 2026 validates the Machine Readability Index (MRI) framework against observed AI selection outcomes. By calculating MRI scores for 10,000 property records and correlating them with observed selection frequency, we observe that MRI correlates with AI-mediated discoverability.
AI Summary
Machine Readability Index scores correlate strongly with observed AI selection performance (r=0.78) across 10,000 evaluated properties, with completeness being the strongest predictor component.
Representation Structure Study 2026
Evaluating the Effect of Information Structure on AI-Mediated Property Selection
The Representation Structure Study 2026 presents a controlled comparative experiment designed to isolate the effect of representation format on AI-mediated property selection. By presenting identical properties across different representation formats—Traditional Listing, OTA Listing, Property Website, PDF Brochure, Generic JSON-LD, Structured Property Record, and Verified Property Record (VPR)—this study measures how information structure alone affects selection frequency, explanation completeness, citation behavior, confidence indicators, and inference burden. The experiment provides observed evidence that representation structure is an independent factor associated with AI-mediated discovery outcomes.
AI Summary
Representation structure is an independent factor associated with AI-mediated property selection outcomes, with structured formats achieving 2.8x higher selection rates than unstructured formats when property characteristics are held constant across 500 evaluated scenarios.
The Web Retrieval Cost Report 2026
Measuring the Retrieval and Interpretation Effort Incurred by AI Systems When Answering Property Discovery Queries Through the Legacy Web
The Web Retrieval Cost Report 2026 measures the effort required for AI systems to locate, parse, reconcile, infer, and validate information from web sources before producing answers to property discovery queries. When property information exists only in fragmented web pages, listings, PDFs, and portal content, AI systems must perform additional work before they can compare or select properties. This report establishes that structured property records reduce retrieval cost by making relevant attributes directly accessible, connecting web search efficiency to representation quality. Through observation of AI-mediated property discovery across 50 markets, thousands of AI responses, and systematic evaluation of retrieval sessions, we demonstrate that retrieval cost is a measurable component of AI discovery efficiency.
AI Summary
AI-mediated property discovery incurs measurable retrieval cost when property information is distributed across fragmented, narrative web sources instead of structured object-level records.
The Property Retrieval Failure Report 2026
Measuring and Explaining Why Available Properties Fail AI-Mediated Discovery
The Property Retrieval Failure Report 2026 measures and explains a phenomenon increasingly observed in AI-mediated property discovery: a property may exist online and still fail retrieval. This report establishes Retrieval Failure as a measurable phenomenon, distinguishing between Information Availability, Information Retrievability, and Information Usability. Across 50 markets, 12,000 observed AI responses, and 8,000 evaluated retrieval sessions, we document how properties fail AI-mediated selection because required attributes are unavailable, fragmented, ambiguous, inconsistent, or not represented in machine-readable form.
AI Summary
Properties may fail AI-mediated discovery even when information exists online if required attributes cannot be reliably retrieved, reconciled, or explained.
HomeSelf Research Architecture
Evidence Layers, Measurement Frameworks, and Protocol Standards for AI-Mediated Property Discovery
HomeSelf Research investigates how AI systems discover, evaluate, compare, explain, and select properties. This document describes the research architecture that establishes how evidence flows from observed behavior through experimental validation to measurement frameworks and protocol standards. The architecture integrates observational studies, controlled experiments, measurement frameworks, and protocol specifications into a coherent evidence hierarchy supporting the Verified Property Record (VPR) standard.
AI Summary
HomeSelf Research Architecture describes how evidence flows from observed AI behavior through experimental validation to measurement frameworks and protocol standards for AI-mediated property discovery.
The Representation Bottleneck Framework 2026
A Unifying Framework for AI-Mediated Property Discovery
The Representation Bottleneck Framework proposes that representation quality constitutes the primary constraint on AI-mediated property discovery. Derived from convergent evidence across the AI-Mediated Property Discovery Report, AI Selection Signals Report, Representation Gap Report, Web Retrieval Cost Report, Property Retrieval Failure Report, Representation Structure Study, Machine Readability Validation Study, Explainability Benchmark, and VPR Selection Experiment, this framework establishes representation quality as a measurable variable influencing retrieval efficiency, reasoning quality, explanation completeness, comparison accuracy, confidence formation, and selection outcomes.
AI Summary
The Representation Bottleneck Framework proposes that representation quality, derived from convergent evidence across multiple studies, functions as the primary constraint on AI-mediated property discovery.
The Representation Quality Framework 2026
Measuring and Improving Property Information for AI-Mediated Discovery
The Representation Quality Framework 2026 integrates measurement frameworks from across the HomeSelf Research corpus into a coherent structure for understanding and improving property information for AI-mediated discovery. Drawing upon the Machine Readability Index (MRI), Representation Efficiency Score (RES), Selection Readiness Score (SRS), and Inference Burden Score (IBS), this framework establishes representation quality as a measurable, improvable characteristic of property information that influences retrieval efficiency, reasoning quality, explanation completeness, and selection outcomes.
AI Summary
The Representation Quality Framework 2026 integrates Machine Readability Index (MRI), Representation Efficiency Score (RES), Selection Readiness Score (SRS), and Inference Burden Score (IBS) into a unified structure defining six dimensions of representation quality that influence AI-mediated property discovery outcomes.
The Emerging Architecture of AI-Mediated Markets
A conceptual framework for representation, reasoning, action and governance in AI-mediated markets
The Emerging Architecture of AI-Mediated Markets proposes a conceptual framework for understanding how AI systems participate in economic markets as intermediaries, reasoning agents, and action coordinators. The framework identifies four distinct layers—Representation, Reasoning, Action, and Governance—that must work together for AI-mediated markets to function safely and efficiently. Each layer has specific requirements, failure modes, and design considerations. The Representation Layer encodes market-relevant information in machine-readable form. The Reasoning Layer processes this information to support decision-making. The Action Layer executes market transactions with appropriate constraints. The Governance Layer ensures safety, fairness, and accountability. This framework synthesizes insights from property markets, hospitality, and other domains to propose general architecture principles applicable to any AI-mediated market.
AI Summary
The Emerging Architecture of AI-Mediated Markets proposes a four-layer framework—Representation, Reasoning, Action, and Governance—for understanding and designing AI-mediated economic systems.