Abstract
AI systems that recommend, refer, or select businesses on behalf of users operate on data never designed for selection decisions. Marketing copy, written to persuade human visitors, and SEO metadata, structured for search engine crawlers, are the primary data sources available to AI assistants making business recommendations. Neither format expresses selection eligibility, capability boundaries, or explicit exclusions.
This paper introduces Deeprank, an open specification that defines a structured declaration format — the Deeprank Selection Profile — for expressing business identity, capabilities, fit conditions, and exclusions. Deeprank addresses the selection layer of AI-mediated discovery, which is architecturally prior to influence optimization techniques such as Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO).
The specification's primary conceptual contribution is the treatment of negative capability — explicit declarations of what a business does not do — as a first-class structural element equal in importance to capability declarations.
Publication Details
- Title
- The Selection Layer: Structured Eligibility Declarations for AI-Mediated Business Discovery
- Author
- Zhgenti, Bakar
- Published
- February 14, 2026
- Publisher
- Zenodo
- License
- CC-BY-4.0
- Version
- 1.0
Key Contributions
Selection Layer Architecture
Identifies and formalizes the selection layer as architecturally prior to influence techniques (AEO, GEO). Proposes a four-layer AI optimization stack where selection eligibility is the foundational gate.
Negative Capability as First-Class Element
Treats explicit declarations of what a business does not do as structurally equal to capability declarations. Exclusions are not an afterthought but a core mechanism for preventing incorrect selection.
Deeprank Selection Profile
A six-layer structured JSON declaration format covering identity, problem declaration, capability, fit conditions, non-fit/exclusions, and stability metadata. Published with a JSON Schema and controlled vocabulary.
Structural Failure Analysis
Demonstrates through concrete scenarios how AI systems making selection decisions from marketing content produce predictable, repeatable errors that structured declarations resolve.
Cite This Work
Zhgenti, B. (2026). The Selection Layer: Structured Eligibility Declarations for AI-Mediated Business Discovery. Zenodo. https://doi.org/10.5281/zenodo.18641029