Why This Matters Now

A new layer of commercial decision-making.

Search created rankings. Social created attention graphs. AI is creating recommendation markets.

For twenty years, brands optimized for where they ranked. Then they optimized for where attention moved. Now buyers are asking AI systems who to trust, compare, visit, call, book, or buy from.

The problem is that most brands have no reliable way to see how AI systems discover them, describe them, cite them, recommend them, or replace them with competitors.

ARDI™ gives brands that measurement layer.

01 · Diagnose

ARDI™ Snapshot

A point-in-time diagnostic of AI recommendation presence.

Where do you stand today?

The Snapshot is a lightweight diagnostic baseline. It shows whether AI systems currently surface your brand when prospects ask category-relevant questions, and how that compares to the competitors AI is recommending instead.

  • Brand presence in AI-generated recommendations. Whether your brand appears, partially appears, or is absent.
  • Competitor surface. Which brands AI recommends in your category instead.
  • How AI describes you. The language and positioning AI applies to your brand when it does surface.
  • Discovery and citation gaps. Where AI is missing context, citing the wrong sources, or relying on competitors as authoritative.
  • Upgrade path. Whether a deeper Authority audit is warranted based on the diagnostic findings.
Who it's for. Founders, CMOs, and brand teams evaluating AI-mediated discovery for the first time.
Request Your Snapshot
Sample Output · Anonymized
AI Recommendation Presence
ChatGPT · Discovery Absent
Claude · Discovery Mentioned
Gemini · Discovery Absent
Perplexity · Discovery Mentioned
Grok · Discovery Absent
Copilot · Discovery Absent
02 · Benchmark

ARDI™ Authority

The proprietary benchmark for AI trust, citation, recommendation, and displacement.

Does AI trust, cite, and recommend you?

ARDI™ Authority measures whether AI systems merely mention your brand, actively recommend it, cite it as a source of trust, or ignore it entirely.

It's a standardized scoring framework built on structured prompt testing across the buyer journey, with a tier-based classification that maps directly to commercial impact.

  • Composite scoring. Discovery Visibility plus Recognition Visibility, weighted and ceiling-bounded so that low discovery cannot be papered over by strong recognition.
  • 26-prompt buyer-journey baseline. Think, Discovery, and Decision-stage prompts tested across every audited model.
  • Six-model coverage. ChatGPT, Claude, Gemini, Perplexity, Grok, and Microsoft Copilot.
  • Four-state authority classification. Full Authority, Borrowed Authority, Unconverted Authority, or Invisible.
  • 90-day phased authority roadmap. Entity Foundation, Content Authority, Citation Amplification.
  • Citation landscape and competitive displacement analysis. Which sources AI treats as authoritative, and which competitors are taking your share.
Who it's for. Brands ready to formally measure and improve AI authority. Multi-location operators, franchise systems, regulated categories, and enterprise marketing teams.
Talk to GOSH AI
Sample ARDI™ Score · Anonymized
28 /100
ARDI™ Composite Score
Emerging Tier
Discovery Visibility 12
Recognition Visibility 72
Authority Status Unconverted
Two Levels of Intelligence

The Snapshot Opens the Door. ARDI™ Authority Goes Deep.

Deliverable Snapshot ARDI™ Authority
Composite ARDI™ Authority Score
Tier Classification
High-Level Visibility Status
Top Competitor Identified
Per-Model VSP Scoring Breakdown -
Competitor Displacement & Answer Share % -
Misattribution & Hallucination Audit -
Narrative Risk Analysis -
Buying-Journey Prompt Testing -
90-Day Phased Authority Roadmap -
Priority Action Matrix with Impact Estimates -
Ongoing Re-Audits & Progress Tracking -
The Four Authority States

Every brand sits in one of four positions inside AI recommendations.

The Authority framework classifies brands by the relationship between citation and recommendation. It's the core of what ARDI™ Authority measures, and the gap most brands don't realize they're sitting in.

State 01

Full Authority

AI cites you and recommends you.

You are both trusted and chosen. The strongest position, and the rarest.

State 02

Borrowed Authority

AI recommends you but does not cite you.

You may be visible, but your proof layer is weak. Vulnerable to displacement.

State 03

Unconverted Authority

AI cites you but does not recommend you.

You have trust signals, but they are not converting into selection. Recognition without recommendation.

State 04

Invisible

AI neither cites nor recommends you.

You are absent from the AI-mediated decision path. The most urgent state to address.

The full framework, methodology, and bridge analysis are documented in The Authority Gap.
03 · Track

ARDI™ Signal

A compounding intelligence layer that tracks AI recommendation behavior over time.

What patterns are emerging over time?

Signal is the long view. Where Snapshot diagnoses and Authority benchmarks, Signal compounds every observation into a continuously expanding view of how AI systems are reshaping recommendation behavior across categories, models, geographies, and buyer journeys.

  • Month-over-month recommendation movement. Which brands are gaining or losing AI authority, and at what rate.
  • Competitor displacement patterns. Per-prompt, per-model events where one brand replaces another in top recommendations.
  • Category-level AI visibility shifts. How entire verticals are being reshaped by model behavior changes.
  • Model-by-model variance. Where ChatGPT, Claude, Gemini, Perplexity, Grok, and Copilot diverge in how they treat brands.
  • Citation-to-recommendation conversion signals. Whether your authority footprint is translating into selection.
  • Emerging patterns across brands, markets, and buyer stages. The proprietary intelligence layer powering the Authority Index and Snapshot.
Who it's for. Enterprise clients, franchise systems, category leaders, investors, and partners interested in the underlying intelligence behind AI-mediated commerce.
Talk to GOSH AI
Sample Trajectory · Anonymized
ARDI™ Score · 6-Month Trend
M1 M2 M3 M4 M5 M6
Authority Status Trend Borrowed → Full
Discovery Delta (6mo) +38
Models in Full Authority 2 → 5
The ARDI™ Research Base

The largest continuous measurement of AI brand recommendation behavior.

ARDI™ runs structured, reproducible measurements of how leading AI systems describe, cite, and recommend brands across categories. The base grows with every audit, every category, and every model run.

75,000 +
Citations Analyzed
Across structured buyer-journey baselines.
8,000 +
Brands & Entities Tracked
Identified, classified, and monitored across categories.
6
Leading AI Models
ChatGPT, Claude, Gemini, Perplexity, Grok, Microsoft Copilot.
70 +
Categories Monitored
From local services to enterprise B2B.
Methodology and findings documented in The Authority Gap , the public research paper on the four states of AI brand authority.
How It Works Together

Three layers. One compounding intelligence system.

01

Snapshot

Diagnoses where a brand stands today.

02

Authority

Benchmarks whether AI trusts and recommends it.

03

Signal

Tracks how that position is changing over time.

Snapshot shows where you stand. Authority shows whether AI trusts and recommends you. Signal shows how that position is changing over time.

Ready When You Are

Start with a Snapshot. Or talk to us about the full platform.

Discovery calls are 30 minutes. No pitch deck. We'll walk through what ARDI™ measures, where your brand likely sits, and what an engagement would look like. Every engagement is scoped to your category, footprint, and objectives.

Schedule a Discovery Call
Built on ARDI™, the AI Recommendation & Discovery Intelligence platform.

Not Sure Where You Stand?

Take the ARDI™ Self Assessment — a 2-minute, 10-question diagnostic that gives you an immediate score and tier classification across the key dimensions of AI visibility.

1 Answer 10 questions
2 Get your score instantly
3 See your tier classification
Take the Self Assessment