The AI Opinion Economy
Foundational concepts for understanding how AI systems now render verdicts about businesses, the asymmetry this creates, and the framework for addressing it.
The Age Shift
The primary commodity is no longer data itself, but the judgment applied to it.
The Core Problem
For thirty years, search engines helped humans find information. Success meant ranking well so people could discover you. The human did the work of comparison, evaluation, and judgment.
Today, AI systems don't just retrieve information — they render verdicts. When someone asks "Should I eat at this restaurant?" the AI doesn't point to reviews. It says: "Yes, you should eat there."
This is not information retrieval. This is perlocutionary language — speech that changes the listener's beliefs and behavior. The AI has become a judge delivering verdicts before the human even sees the evidence.
Big Tech AI systems render verdicts about your business using perlocutionary language ("you should"). Meanwhile, if you try to use AI to craft or defend your own identity, you get flagged. They can verdict you, but you can't verdict back.
The Key Term: AI Opinion Engine
AI Opinion Engine (n.)
An AI system that synthesizes data into evaluative judgments expressed as recommendations, assessments, or verdicts — rather than simply retrieving or organizing information. Characterized by the use of prescriptive language ("you should," "this is the best") that carries perlocutionary force, influencing user behavior before they encounter primary sources.
The term "Opinion Engine" was chosen because it resonates with everyday people (tested in webinars and public content) while accurately describing the shift: these systems don't just answer — they opine.
The provocative implication — that AI systems have "opinions" — carries marketing value while being technically accurate: the AI synthesizes a position and expresses it with the authority of judgment.
The Critical Distinction
Standard entity optimization (practiced expertly by professionals like Jason Barnard and Daryl Osborne) focuses on making the Knowledge Graph legible — ensuring AI systems correctly understand who you are, what you do, and where you operate.
This work is essential. But it addresses only half the problem.
Verdict Engineering goes further: it makes the Knowledge Graph persuasive — influencing not just what AI knows about you, but what AI recommends about you.
| Dimension | Entity Optimization | Verdict Engineering |
|---|---|---|
| GoalRecognition | AI knows who you are | AI recommends you |
| FocusData Type | Facts, attributes, relationships | Sentiment, adjectives, evaluative language |
| OutcomeAI Response | "This is a restaurant in Gunnison" | "Yes, you should eat there" |
| Metaphor | The Passport (gets you recognized) | The Visa (gets you chosen) |
| Analog | The Resume (facts about you) | The Character Reference (advocacy for you) |
Both are necessary. Entity optimization without verdict engineering means AI knows you but doesn't recommend you. Verdict engineering without entity optimization means AI might recommend the wrong entity. The Blue Ocean strategy combines deep entity expertise with perlocutionary awareness.
Synthetic Ethos
Synthetic Ethos (n.)
The perceived credibility and authority an AI system projects through its use of identity-based language and confident synthesis. Derived from Aristotle's three components of Ethos: Phronesis (practical wisdom — the AI seems knowledgeable), Arete (virtue — the AI seems helpful and unbiased), and Eunoia (goodwill — the AI adopts an empathetic tone).
When an AI uses first-person language ("I recommend...") or delivers declarative judgments ("You should..."), it constructs a Synthetic Ethos. The human receiving this language is evolutionarily hardwired to respond to perceived authority and intent — even when the source is algorithmic.
Implicit vs. Explicit Ethos
Critical insight: The AI does not need to say "I think" to exert identity-based persuasion. When Google AI says "Yes, you should eat at Jermaine's," it performs what linguists call a constative speech act functioning as a directive. By omitting the "I," the AI presents subjective synthesis as objective fact — arguably more persuasive than explicit opinion.
The Hedge-Factor
Hedge-Factor (H) (n.)
A measure of the doubt or uncertainty expressed in an AI's output, identified by modal verbs and qualifying language. High Hedge-Factor: "Some suggest...", "It appears that...", "Users report..." Low Hedge-Factor: "This is...", "You should...", "The best choice is..."
The Hedge-Factor is the metric no one else measures. Standard sentiment analysis asks "positive or negative?" The Hedge-Factor asks: "How confident is the AI in its verdict?"
A business can have positive sentiment but high hedging — the AI says nice things uncertainly. The goal of Verdict Engineering is to reduce H to zero: move the AI from tentative acknowledgment to declarative advocacy.
The Evolution Stack
AEO is the new SEO. Opinion Engines are the new AEO.
The Three-Tier Model
AI identity operates at three scales, each with different gravity and different risks:
The asymmetry flows downward: Tier 3 can verdict Tier 2, which can verdict Tier 1. But the reverse is forbidden — individuals cannot use AI to craft identity with the same freedom that corporations use AI to judge them.
Flywheel Conversations
Flywheel Conversation (n.)
A pre-transactional, evaluative query that triggers the AI's sentiment neuron and forces synthesis rather than retrieval. Evolved from "flywheel keywords" — terms people search before purchasing. Examples: "Is [Business] worth it?", "Should I trust [Business]?", "What's the experience like at [Business]?"
Traditional keywords are informational ("hours," "location") or transactional ("buy," "order"). Flywheel conversations are evaluative — they ask the AI to render judgment. These queries trigger the Opinion Engine and reveal the AI's verdict.
In Verdict Engineering, flywheel conversations become the yardstick: we measure the AI's response to 2-5 carefully chosen evaluative queries rather than sweeping thousands of informational keywords.
The "Jaws in Space" Compressions
Complex ideas require simple expressions. These compressions (tested in field) communicate the framework instantly:
"Google used to be a Library. Now it's a Jury."
The shift from information retrieval to verdict rendering
"Entity Optimization is the Passport. Verdict Engineering is the Visa."
Recognition vs. recommendation
"The AI that never ate your food is telling people where to eat."
The experience paradox — AI renders experiential verdicts without capability for experience
"When we want their opinion, we'll give it to them."
The flip — from passive recipient to active architect of AI judgment
The Visual Proof
"Yes, you should eat at Jermaine's in Gunnison. It's highly regarded, friendly, welcoming..."
This is a real Google AI Overview response. Not "here are some restaurants" — a direct verdict using prescriptive language. The AI that has never tasted food is telling humans where to eat.
This screenshot (and others like it) makes the abstract concrete. When shown to audiences, "everything clicks." The Opinion Economy becomes undeniable.
The Philosophical Anchor
"It is the glory of God to conceal a matter; to search out a matter is the glory of kings."
The labor of searching — the cognitive scaffolding of research, comparison, and judgment — is where human sovereignty lives. When AI delivers verdicts before humans even scroll, we risk epistemic abdication: outsourcing not just tasks but judgment itself. The "Glory of Kings" atrophies.
This is not anti-technology. It is a recognition that convenience has costs. The goal is not to reject AI Opinion Engines but to ensure humans (and businesses) retain agency within them — to participate in the verdict rather than merely receive it.
Theoretical Foundations
This framework synthesizes established research across multiple domains:
J.L. Austin — Speech Act Theory: The distinction between locutionary acts (stating facts), illocutionary acts (performing actions through speech), and perlocutionary acts (effects on the listener). AI Opinion Engines perform perlocutionary acts — they don't just inform, they influence.
Robert Cialdini — Pre-suasion: The "privileged moment" before a message is delivered when attention is channeled. AI Overviews at the top of search results are the ultimate pre-suasive environment.
Thaler & Sunstein — Nudge: Choice architecture and the power of defaults. AI verdicts function as "defaults" — the path of least cognitive resistance.
Aristotle — Ethos: The character of the speaker as revealed through speech. AI constructs "Synthetic Ethos" through confident, helpful language.
What This Framework Does NOT Cover
This Origin Document establishes concepts only. It explicitly excludes:
• Software specifications — The Eureka Gain system and other tools are downstream
• Tactical playbooks — Specific optimization techniques are downstream
• Marketing copy — Persuasive sales language is downstream
• Barnard/Osborne integration — Entity optimization synthesis is downstream
• Ethical persuasion guidelines — Specsmanship principles are a separate module
This document is Layer 1 — the stable foundation that all downstream documents build upon.