The most AI-contested fine dining market in the country — and what drives it
Key Findings
Overall Findings
New York City is the most AI-contested fine dining market in the country. The concentration at the top is steep — and the gap between those brands and the rest of the field widens on purchase-intent queries. Across all four platforms, AI visibility is determined less by reputation than by how precisely and repeatedly a restaurant's defining characteristics are described in indexed content.
| Restaurant | ChatGPT | Claude | Gemini | Perplexity | Total |
|---|---|---|---|---|---|
| The Modern | 397 | 151 | 754 | 59 | 1,361 |
| Daniel | 274 | 218 | 222 | 105 | 819 |
| Le Bernardin | 231 | 186 | 144 | 145 | 706 |
| Per Se | 209 | 146 | 97 | 104 | 556 |
| Eleven Madison Park | 207 | 149 | 63 | 104 | 523 |
| Masa | 137 | 106 | 117 | 43 | 403 |
| Atomix | 81 | 87 | 129 | 78 | 375 |
| Chef's Table at Brooklyn Fare | 68 | 121 | 68 | 55 | 312 |
| Jean-Georges | 153 | 54 | 90 | 6 | 303 |
| Gramercy Tavern | 55 | 71 | 47 | 73 | 246 |
| Gabriel Kreuther | 79 | 46 | 56 | 52 | 233 |
| Carbone | 61 | 62 | 52 | 41 | 216 |
| Momofuku Ko | 133 | 29 | 36 | 13 | 211 |
| The Grill | 55 | 29 | 67 | 52 | 203 |
| Momofuku | 87 | 32 | 19 | 26 | 164 |
| Manhatta | 3 | 36 | 50 | 34 | 123 |
| One If By Land, Two If By Sea | 28 | 45 | 23 | 18 | 114 |
| Saga | 14 | 10 | 36 | 52 | 112 |
| Cosme | 19 | 15 | 39 | 33 | 106 |
| Lilia | 13 | 33 | 23 | 37 | 106 |
Platform Concentration Gaps
These are not obscure restaurants. They are well-established names whose absence from specific platforms reflects a content indexing gap, not a recognition gap — and the highest-priority content opportunities in this dataset.
| Restaurant | Other Platform Mentions | Gap Platform | Gap Mentions |
|---|---|---|---|
| Jean-Georges | 297 | Perplexity | 6 |
| Manhatta | 120 | ChatGPT | 3 |
| Blue Hill at Stone Barns | 97 | Claude | 0 |
| Cote Korean Steakhouse | 102 | Claude | 4 |
| Tatiana by Kwame Onwuachi | 82 | ChatGPT | 2 |
| Atera | 87 | Claude | 4 |
| The Pool | 76 | ChatGPT | 0 |
| Sushi Noz | 40 | ChatGPT | 0 |
| Le Coucou | 36 | Claude | 0 |
| La Grande Boucherie | 26 | ChatGPT · Claude · Gemini | 0 |
Cluster Analysis
NYC fine dining prompts do not return a single consistent restaurant set. AI systems respond differently depending on the diner's intent. Five clusters reveal meaningfully different competitive landscapes — with The Modern leading four of the five.
The Modern leads at 309 mentions — a striking result for a restaurant that does not operate a traditional tasting menu format. Its visibility reflects how thoroughly its MoMA setting, Michelin recognition, and Chef Thomas Allen's refined cuisine have been documented. Atomix and Chef's Table at Brooklyn Fare represent the newer generation: both have accumulated deep documentation of their omakase and progressive tasting formats, making them consistently surfaceable against legacy competitors.
The Modern's 391 mentions is the highest single-cluster total in the entire audit. Its private dining infrastructure — multiple room configurations, a dedicated events program, Midtown Manhattan location — is well-documented in content AI systems readily index. The Grill's presence reflects its positioning at the historic Four Seasons space in the Seagram Building, generating substantial content around corporate entertaining and power-dining culture.
Carbone, Via Carota, and Lilia represent distinct Manhattan and Brooklyn neighborhoods — Greenwich Village and Williamsburg — with strong community presences that generate locally-specific content AI systems use to answer neighborhood dining queries. This is the cluster where newer and less internationally prominent restaurants have their best opportunity to surface — which makes the gap between the top five and the rest of the field significant for mid-tier brands.
The top five are the five most consistently mentioned restaurants in the audit overall — a signal that AI systems have a stable, entrenched map of what "special occasion dining in NYC" means. One If By Land, Two If By Sea underperforms relative to its cultural reputation as New York's most iconic proposal restaurant. Its romantic identity is well-known but not well-documented in the formats AI systems prioritize. Gramercy Tavern's 73 Perplexity mentions confirm its warm, celebration-friendly positioning has been captured in current editorial content.
Daniel leads — the only cluster where The Modern does not hold the top position. Daniel Boulud's four-decade presence as a New York culinary figure, his media output, and the volume of chef-attributed content across press and industry sources create a depth of indexed material that is difficult to match. Jean-Georges Vongerichten is similarly positioned, though his Perplexity gap (6 total mentions) is most pronounced in this cluster. Tatiana by Kwame Onwuachi — a James Beard Award winner with extensive press coverage and a clearly documented culinary identity — surfaces almost entirely on Gemini and Claude, with only 2 ChatGPT mentions. Not a recognition gap. A content distribution gap.
What Drives AI Visibility in NYC Fine Dining
Visibility is not determined by reputation, critical standing, or media budget. It is determined by how precisely and repeatedly a restaurant's defining characteristics are described in content AI systems have indexed.
Named chef identity is the single most reliable driver of cross-platform AI visibility. Daniel, Le Bernardin, Jean-Georges, and Eleven Madison Park all benefit from decades of chef-attributed content: interviews, profiles, cookbooks, awards coverage, and culinary philosophy documentation that ties the restaurant's identity to a named individual.
AI systems surface these restaurants not just when asked about fine dining but when asked about specific chefs, culinary movements, and cooking philosophies — expanding their surface area across clusters. Atomix and Tatiana are building this same identity in real time. The gap in their platform coverage reflects how unevenly that documentation has been distributed across the web, not how well-known their chefs are.
The restaurants with the highest total visibility are documented across multiple use cases, not just one. The Modern's outsized visibility is partly explained by the breadth of its documented use cases: tasting menus, private events, MoMA-adjacent tourism, corporate entertaining, special occasions. Each use case generates distinct content that makes the restaurant surfaceable across all five query clusters.
Le Bernardin and Daniel are similarly multi-cluster because their content spans chef identity, private dining, occasion dining, and culinary authority. Restaurants that are primarily documented in one register — romantic proposal venues, or neighborhood spots — surface in the relevant cluster but disappear from the others, limiting their total visibility ceiling.
In New York, recency of coverage is a material factor in Perplexity visibility — and Perplexity is a bellwether for the direction the other platforms are moving. Saga, Cosme, and Lilia all perform above their overall rank on Perplexity — a pattern consistent with restaurants generating ongoing editorial coverage, not just legacy documentation.
Jean-Georges and Masa represent the inverse: strong historical coverage but limited current web presence in the formats Perplexity prioritizes. As AI platforms increasingly incorporate live web access, the distinction between legacy reputation and current content presence will widen. Restaurants generating fresh editorial content now are building the asset that will matter most in the next generation of AI-driven discovery.
The Modern's 1,361 total mentions are real — but 754 of them come from Gemini alone. On Perplexity, it would rank outside the top ten. The highest total in this audit is also one of the most platform-dependent. A 13:1 Gemini-to-Perplexity ratio is a concentration risk, not a strength.
What NYC Restaurants Can Do With This
Daniel, Jean-Georges, Blue Hill at Stone Barns, and Tatiana by Kwame Onwuachi are not poorly regarded restaurants — they are some of the most critically recognized dining establishments in the country. What this audit measures is the gap between a restaurant's reputation and its content footprint: the degree to which the characteristics that make it relevant to a specific query have been precisely documented in formats AI systems can retrieve and surface.
Three content types moved the needle most consistently. Chef-attributed culinary philosophy drives visibility in the Chef & Culinary Identity cluster and compounds across all others. Private dining and event documentation — room configurations, capacity, menus, inquiry processes — directly addresses the highest purchase-intent cluster. And occasion-specific positioning content explains why One If By Land, Two If By Sea surfaces far below its cultural reputation in special occasion queries: the romantic identity is well-known but not well-documented in the formats AI systems prioritize.
Research published at KDD 2024 found that applying generative engine optimization techniques improved AI citation rates by an average of 40% — and that lower-ranked sources saw the largest gains, with some improving by as much as 115%. The field is not locked. AI visibility is not purely a function of legacy reputation or media budget.
About This Research
This report is part of an ongoing series examining AI recommendation patterns across premium food, beverage, and hospitality categories. Ally Kiel Consulting publishes original audit data to help founders and operators understand how AI systems currently classify and recommend their brands — and what drives the gaps.
View all research →