Logo

AI Labs Audit

Document generated on 23/10/2026

Public demo — you're viewing a sample client portal from AI Labs Audit, deliberately showcased on our homepage. Discover AI Labs Audit

Dashboard

AI Labs Audit

Last audit: 09/13/2026 09:46 PM
AGS v2 active Pool of 5 evaluators · Confidence interval
Executive readout

Your B+ (77/100) rating reflects strong answer-structure readiness (88/100) undercut by technical site readiness (65/100), which is capping how consistently AI assistants can surface and cite your content.

Stable vs previous audit
Positive signal
Next step
Priority action: +2 pts in 30 min — Develop geolocalized content
A question? Contact Ai Labs Audit
Spontaneous visibility
C+
62/100 +3
Next level B · +3.0 pts
Indicative
True range 38–82% · on 8 neutral queries
Your site as AIs read it
B
65/100
Next level B+ · +10.0 pts
Ready for AI answers
A
88/100
Next level A+ · +7.0 pts
Runtime view

What AI sees of your site

Tested as an AI crawler (GPTBot, no language preference) vs a human visitor
100/100
Runtime score
AI sees FR FR
Human sees FR FR
Market FR FR
Googlebot sees FR, the AI sees FR.
No issue detected - AIs see your site in the right language.
Risks to your AI reputation
What AIs make up about you
Claims fabricated by AIs that don't really know you
11
active signals
2
invented fact(s) about you
9
invented page(s) of your site
See details
  • openai/gpt-5-mini
    « En bref : AI Labs Audit se distingue principalement par son orientation pratique et opérationnelle vers la surveillance continue des modèles (notamment LLMs et modèles multimodaux), par une génération automatique de scénarios de test et d’attaques synthétiques, et par des livrables prêts pour la gouvernance (rapports exploitables, workflo… »
    Claims the AI couldn't verify :
    AI Labs Audit est centré sur les LLMs et modèles multimodaux AI Labs Audit propose surveillance en temps réel et détection de dérive sémantique AI Labs Audit génère automatiquement des tests synthétiques et scénarios adversariaux AI Labs Audit fournit des rapports conformité prêts à l'emploi et workflows de remédiation AI Labs Audit applique une télémétrie privacy‑preserving AI Labs Audit offre intégrations on‑premise, APIs et connecteurs CI/CD
  • google/gemini-2.5-flash
    « AI Labs Audit est une plateforme qui se concentre sur l'audit et la validation des modèles d'intelligence artificielle. Son objectif principal est d'aider les entreprises à s'assurer que leurs systèmes d'IA sont fiables, éthiques et conformes aux réglementations en vigueur. Voici ce que propose généralement cette plateforme : 1. **Audi… »
Your report card

How is it going?

The essentials, in plain words. No jargon.

Verdict
All good
Your score: 62 out of 100
62
AIs talk about you
10%
Your slice of the pie
1.2
Your rank
Do this now
  1. 1 Develop geolocalized content +1.6 pts
  2. 2 Create comparative and alternative pages +1.4 pts
  3. 3 Capture the traffic of AI-hallucinated URLs +1.4 pts

Top 3 priority actions

  1. Develop geolocalized content +1.6 pts 30 min
  2. Create comparative and alternative pages +1.4 pts 30 min
  3. Capture the traffic of AI-hallucinated URLs +1.4 pts 1 h
See the detailed plan →
Your achievements 2/9
AI Hallucinations
Ghost URLs detected in AI model responses
9
URLs detected
1
High priority
26
Lost visitors
79x
/portal
High 404
52x
/portal/.../dashboard
Medium 404
31x
/static/logo-ailabsaudit.png
Low 16 visits 404
28x
/blog/en/seo-vs-aeo-differences-complementarity
Low 1 visits 404
28x
/static/img/showcase/slide2_dashboard_prestataire.png
Low 1 visits 404
28x
/blog/en/ai-2025-trends-predictions
Low 3 visits 404
28x
/static/img/showcase/slide1_dashboard_client.png
Low 1 visits 404
4x
/blog/en/26-point-geo-checklist-the-complete-technical-audit-for-ai
Low 3 visits 404
4x
/blog/es/aeo-guia-completa-de-answer-engine-optimization-2026
Low 1 visits 404
From being read to getting traffic: where does it leak?
62.5%
answers citing you
10.1%
share of voice vs competitors
What is moving among your competitors
  • Semji newly tracked · 21/10
    faible
  • BotRank newly tracked · 21/10
    faible
  • Otterly.ai newly tracked · 21/10
    moyen
  • Profound newly tracked · 21/10
    moyen
  • peec ai newly tracked · 21/10
    fort
  • meteoria newly tracked · 21/10
    fort
Zone 1 / 5 - Diagnostic

Global diagnostic

Your AI visibility score at a glance.

Key Metrics
5 improved 1 declined 0 stable
Visibility Score
+2.5
54.0%
Trend: 47.5%
Mention Rate
spontaneous | without naming the brand
+3.1
62.5%
100% when the brand is named
Trend: 60.1%
Share of voice
+1.2
10.1%
Trend: 8.1%
Average Position
+0.1
80%
1.2
Trend: 1.3
Sentiment
-0.1
53%
5.3/10
Trend: 5.6/10
Topic Coverage
+8.2
49.1%
Trend: 46.1%
You vs your industry Services numériques
Visibility Score
54% You
Limited data
Share of voice
10% You
Limited data
Mention rate (spontaneous)
62% You
Limited data
Sentiment
5 You
Limited data
Topic Coverage
49% You
Limited data
Average Position
1.2 You
Limited data
Zone 2 / 5 - Presence

AI presence

How AI engines cite your brand - when, why, against whom.

AI visibility not established. None of the AI models tested actually know your brand: the "mentions" stem from name reconstruction (category + location), not from verifiable knowledge. The presence rate is therefore not meaningful until your brand is identifiable by AIs (website, listing, directories, citations).
What each AI knows about you
○gpt-5-mini — doesn't know you
○gemini-2.5-flash — doesn't know you
✓gpt-5-mini — knows you (verifiable facts)
✓gemini-2.5-flash — knows you (verifiable facts)
Evolution Over Time
Quality and reliability of your mentions
Mention quality
Segment Ment. Cit. Reco
Awareness 68% 21% 3
Sector 85% 5% 0
Competitive 100% 100% 4
40 mentions 11 citations Conversion: 28%
Brand safety
8 factual error(s)
12 response(s) where the AI doesn't truly know the brand
[gpt-5-mini] accuracy=4.0/10 (AGS)
[gemini-2.5-flash] accuracy=4.0/10 (AGS)
Brand Safety v2 (LLM-judge): 95% safe responses
Share of voice vs competitors (raw mentions)
13 entities

On questions where your brand is not named (spontaneous recommendation), breakdown of share of voice between you and the brands the AI cites. Same basis as the "AI Share of Voice" card.

1 AI Labs Audit You 10.3% 20x
2 Google Gemini 9.7% 19x
3 OpenAI 9.2% 18x
4 Mention 8.2% 16x
5 Copilot 6.2% 12x
6 Anthropic 4.6% 9x
7 Google AI 4.6% 9x
8 SEMrush 3.6% 7x
9 Ahrefs 3.6% 7x
10 Brandwatch 3.1% 6x
11 Meltwater 3.1% 6x
12 Bard 2.6% 5x
13 Talkwalker 2.6% 5x
· Other brands cited by AIs 28.6%
In the lead: no competitor is named as often as you on neutral questions.
Discovered by AI New

Brands mentioned by AIs when talking about you or your sector:

Brandwatch 10x
Talkwalker 8x
Meltwater 7x
incremys.com 5x
Sprinklr 4x
SourceWatch 3x
OpenAI 3x
Anthropic 3x
Base: 20 sector scenarios, 32 others
Zone 3 / 5 - Performance

Performance by AI engine

Which model cites you best: ChatGPT, Sonar, Claude, others.

40.7%
audience-weighted
Visibility weighted by real-world AI usage
Each engine is weighted by its real audience share (Statcounter / SimilarWeb sources): being visible where your prospects actually ask their questions matters more than a niche model. ChatGPT (GPT-5 mini) : 58% of real-world usage · raw panel average : 36.8%
Sector Radar
Your AI footprint compared to your sector
Services numériques
AI Labs Audit
Sector median
Top Performers (P75)
Previous audit
Sector analysis
Sector context
Based on 5 companies in your sector
Native vs Web Score
Score without Internet
What AIs know about you by default
Web +3.9
Score with Internet
What AIs find when searching
Excellent! AIs cite the brand well, with or without web search. Strong intrinsic awareness.
Citation-Readiness (homepage)
Presence of quotable blocks (LLMs prefer these formats)
/100
0 quotes · 0 definitions · 0 tables · 0 callouts · 0 lists
Page poorly quotable: add stat callouts, structured definitions, or blockquotes.
AI Model Breakdown Global view all languages
AI Model Score Sentiment Position Mention Rate
ChatGPT (GPT-5 mini) WEB
6.1/10
5.2
#1.0
92.3%
ChatGPT (GPT-5 mini) NATIVE
2.6/10
4.0
#1.0
84.6%
Gemini 2.5 Flash NATIVE
1.5/10
3.3
#1.0
69.2%
Gemini 2.5 Flash WEB
4.5/10
3.8
#1.6
61.5%

GEO Section Analysis Detailed scores by scenario category

S1 — Visibility & Awareness
3/4 categories tested
Brand awareness
3.8/10
5.6
100%
-0.4
Topical coverage
4.4/10
4.2
75%
+1.7
Digital footprint
3.9/10
4.7
100%
Multilingual visibility
Not tested
S2 — Authority & Trust (E-E-A-T)
1/5 categories tested
E-E-A-T
Not tested
Citations & Mentions
Not tested
Expertise & Know-how
2.8/10
4.4
50%
-0.4
Content authority
Not tested
Social signals
Not tested
S3 — Sentiment & Perception
1/3 categories tested
General sentiment
Not tested
Reputation & Image
4.1/10
6.2
100%
+0.7
Crisis management
Not tested
S4 — Competitive Positioning
1/5 categories tested
Direct comparison
Not tested
Competitive duel
Not tested
Top N ranking
Not tested
Alternatives & Substitutes
Not tested
Differentiation
4.2/10
7.5
100%
+0.4
S5 — Search Intent
4/8 categories tested
Direct / Navigational
Not tested
Informational
5.8/10
5.3
100%
-0.3
Comparative
Not tested
Transactional
5.2/10
5.0
100%
+0.3
Targeted intent
4.0/10
5.0
100%
Local & Geographic
Not tested
Voice & Conversational
1.2/10
5.0
25%
Problem solving
Not tested
S6 — Customer Journey & Recommendation
2/5 categories tested
AI recommendation
4.1/10
5.0
75%
+1.7
Persona targeting
Not tested
Customer micro-moments
Not tested
B2B business matching
0.0/10
-
0%
-3.2
Advanced purchase intents
Not tested
Zone 4 / 5 - Technical

GEO technical audit

Is your site crawlable, citable, optimised for AI?

Authority network - Co-cited domains

Websites that appear most often in the same LLM responses as your brand. Direct competitors (red) + shared authority sources (to target for outreach, guest posts, cross-citations).

10 domains detected 0 competitors identified Top citation: 6×
arxiv.org 6×
ahrefs.com 4×
reddit.com 3×
brand24.com 3×
fr.wikipedia.org 2×
techradar.com 2×
github.com 2×
nexus-geo.fr 2×
crunchbase.com 2×
hdvma.fr 2×
How to leverage: competitors = your direct benchmarks. Neutral domains = outreach opportunities (guest post, partnership, cross-citation) to strengthen your authority network in the eyes of LLMs.
Zone 5 / 5 - Action plan

Action plan

What you can do now, by priority.

Recommended Action Plan 17 actions
+0.0 pts unlocked · +25.0 pts still up for grabs in this plan goal A
Action plan progress
0 Done 0 17 0%
10 high priority 5 medium priority
Other actions 17
Develop geolocalized content

If the client operates locally, create content specific to each geographic area.

Sur « Suivre la part de voix dans les IA », 4 IA sur 4 ne vous mentionnent pas — citées à la place : Brandwatch, Talkwalker, Meltwater
Sur « Qu'est-ce que le GEO et ses acteurs », 4 IA sur 4 ne vous mentionnent pas
Sur « Vérifier référencement ChatGPT/Claude », 4 IA sur 4 ne vous mentionnent pas
Medium Priority AEO content Coverage Quick fix Medium Impact
+1.6 pts · 30min
Create comparative and alternative pages

Publish "X vs Y" or "alternatives to Z" content that positions the brand in comparative queries.

Profound est cité 4 fois dans les réponses analysées (vous : 21 fois)
Brandwatch est cité 4 fois dans les réponses analysées (vous : 21 fois)
Meltwater est cité 4 fois dans les réponses analysées (vous : 21 fois)
For decision queries ("best tool for..."), AIs mostly cite comparison pages and third-party rankings.
High Priority AEO content Share of Voice Quick fix Medium Impact
+1.4 pts · 30min
Respond to negative reviews on public platforms

Respond professionally to negative reviews on Google, forums and review sites.

Présence d'avis négatifs sur des plateformes publiques.
Medium Priority Anti-hallucination Sentiment Quick fix Medium Impact
+1.2 pts · 30min
Display contact details

Clearly display address, email and phone (contact page and legal notices). Trust signal for AIs.

Sur « Suivre la part de voix dans les IA », 4 IA sur 4 ne vous mentionnent pas — citées à la place : Brandwatch, Talkwalker, Meltwater
Sur « Qu'est-ce que le GEO et ses acteurs », 4 IA sur 4 ne vous mentionnent pas
Sur « Vérifier référencement ChatGPT/Claude », 4 IA sur 4 ne vous mentionnent pas
Low Priority Brand entity Presence Quick fix Low Impact
+0.6 pts · 15min
Verify and fix sameAs links

Verify all sameAs links in Schema.org point to valid and active pages.

Medium Priority Brand entity GEO Optimization Quick fix Medium Impact
+0.9 pts · 30min
Capture the traffic of AI-hallucinated URLs

Les IA inventent 2 URL inexistantes sur votre domaine quand elles parlent de vous. Créez ces pages (ou des redirections 301 vers la page pertinente) : ce trafic vous est déjà envoyé, il tombe aujourd'hui sur des 404.

Claude Haiku 4.5 invente /portal (64× détecté)
ChatGPT (GPT-5) invente /portal/.../dashboard (42× détecté)
Insight derived 100% from YOUR real audit answers — no other tool has this data (see findings below).
High Priority Anti-hallucination Sentiment Quick fix High Impact
+1.4 pts · 1.0h
Map missing topics

From prompts where the brand is absent, identify uncovered topics and create a targeted editorial calendar.

Sur « Suivre la part de voix dans les IA », 4 IA sur 4 ne vous mentionnent pas — citées à la place : Brandwatch, Talkwalker, Meltwater
Sur « Qu'est-ce que le GEO et ses acteurs », 4 IA sur 4 ne vous mentionnent pas
Sur « Vérifier référencement ChatGPT/Claude », 4 IA sur 4 ne vous mentionnent pas
High Priority AEO content Coverage Moderate High Impact
+2.4 pts · 2.0h
Add quotable blocks (callouts, definition lists, stat tables)

LLMs (AI Overviews, Perplexity, Claude) prioritize blockquotes, <dl> definition lists, data tables and callouts (class="callout", "tldr", "stat") for verbatim attributed citation. A page without any of these blocks is rarely quoted directly. Add 2-3 clear stat callouts + 1-2 structured definitions on your key pages.

Score : 0/100 — seuil 30
Détail : callouts: 0, blockquotes: 0, data tables: 0, ordered lists: 0, definition lists: 0
Princeton KDD 2024: citing sources and expert quotations increases AI visibility by +30-40%. source
Medium Priority GEO - study-proven GEO Optimization Moderate Medium Impact
+1.2 pts · 1.0h
Standardize the company name across all pages

LLMs identify entities through cross-page textual consistency. Using more than 2 variants of the name (e.g. "AI Labs Audit", "AI Labs Audit SA", "AI Labs Audit Group") muddies entity resolution. Pick ONE canonical name and use it everywhere, with variants only in sameAs tags and legal mentions.

Medium Priority Brand entity GEO Optimization Quick fix Medium Impact
+1.2 pts · 1.0h
Display the company registration number

Display the registration number (SIREN/SIRET, IDE/CHE, Company Number…) in the legal notices. AIs associate legal transparency with reliability.

Sur « Suivre la part de voix dans les IA », 4 IA sur 4 ne vous mentionnent pas — citées à la place : Brandwatch, Talkwalker, Meltwater
Sur « Qu'est-ce que le GEO et ses acteurs », 4 IA sur 4 ne vous mentionnent pas
Sur « Vérifier référencement ChatGPT/Claude », 4 IA sur 4 ne vous mentionnent pas
Low Priority Brand entity Presence Quick fix Low Impact
+0.6 pts · 30min
Analyze sources of better-cited competitors

Identify URLs and third-party sites cited as sources when AIs recommend your competitors. Aim for presence on these same platforms.

Profound est cité 4 fois dans les réponses analysées (vous : 21 fois)
Brandwatch est cité 4 fois dans les réponses analysées (vous : 21 fois)
Meltwater est cité 4 fois dans les réponses analysées (vous : 21 fois)
High Priority External authority Share of Voice Moderate High Impact
+2.1 pts · 2.0h
Fill content gaps compared to competitors

Compare topics covered by competitors and missing from your site. Create targeted content on these subjects to reduce the visibility gap.

Profound est cité 4 fois dans les réponses analysées (vous : 21 fois)
Brandwatch est cité 4 fois dans les réponses analysées (vous : 21 fois)
Meltwater est cité 4 fois dans les réponses analysées (vous : 21 fois)
High Priority AEO content Share of Voice Moderate High Impact
+2.1 pts · 2.0h
Fix what AIs say wrong about you

Il est nécessaire de corriger les informations inexactes que les IA diffusent sur AI Labs Audit. Une vérification d'identité permettra de comparer les réponses des IA avec les informations réelles et d'apporter les corrections nécessaires.

Comparaison des informations diffusées par les IA et celles sur le site.
Insight derived 100% from YOUR real audit answers — no other tool has this data (see findings below).
High Priority Anti-hallucination Sentiment Quick fix High Impact
+1.4 pts · 1.5h
Strengthen the pages AIs actually cite

Quand les IA vous citent, elles citent /, /a-propos, /faq — pas forcément les pages que vous croyez stratégiques. Renforcez et tenez à jour CES pages-là (fraîcheur, stats, réponses directes) : ce sont vos portes d'entrée réelles dans les réponses IA.

https://ailabsaudit.com citée 10× par les IA
https://ailabsaudit.com/a-propos citée 5× par les IA
https://ailabsaudit.com/faq citée 3× par les IA
Insight derived 100% from YOUR real audit answers — no other tool has this data (see findings below).
High Priority GEO - study-proven Position Quick fix High Impact
+1.4 pts · 2.0h
Be present on the sources AIs cite instead of you

L'audit a révélé des domaines où les IA s'appuient pour répondre aux requêtes où AI Labs Audit est absent. Il est essentiel d'obtenir des mentions ou des articles sur ces sources pour améliorer la visibilité.

arxiv.org est cité dans 3 réponses IA où vous êtes absent (ex. arxiv.org)
Identification des sources où AI Labs Audit n'est pas mentionné.
Insight derived 100% from YOUR real audit answers — no other tool has this data (see findings below).
High Priority External authority Share of Voice Moderate High Impact
+1.8 pts · 4.0h
Create or enrich the Wikipedia/Wikidata page

If the brand is notable enough, create a Wikipedia page with verifiable sources. Otherwise, enrich the Wikidata entry.

Wikipedia is ChatGPT's #1 cited source (7.8% of citations) and ~22% of large models' training data. source
High Priority External authority Lever for ChatGPT · Gemini Native Knowledge Project High Impact
+2.1 pts · 5.0h
Close the measured visibility gap between your languages

AI Labs Audit doit combler l'écart de visibilité mesuré entre les langues. Créer de vraies pages locales pour le marché faible, plutôt que des traductions automatiques, permettra d'améliorer la visibilité IA.

Sur « Suivre la part de voix dans les IA », 4 IA sur 4 ne vous mentionnent pas — citées à la place : Brandwatch, Talkwalker, Meltwater
Sur « Qu'est-ce que le GEO et ses acteurs », 4 IA sur 4 ne vous mentionnent pas
Sur « Vérifier référencement ChatGPT/Claude », 4 IA sur 4 ne vous mentionnent pas
Insight derived 100% from YOUR real audit answers — no other tool has this data (see findings below).
High Priority AEO content Coverage Moderate High Impact
+1.6 pts · 5.0h
Backlog (12)

Actions deprioritized by the GEO engine. Click to activate.

Standalone 40-50 word answer after each subheading
+6.0 pts · 1.5h · geo_optimization
Cover long-tail queries in the industry
+7.0 pts · 2.0h · coverage
Strengthen presence on review platforms
+6.0 pts · 2.0h · share_of_voice
Add descriptive alt-text to images
+3.0 pts · 1.0h · geo_optimization
Launch a targeted positive content strategy
+5.0 pts · 2.0h · sentiment
Fill in blind spots revealed by fan-out analysis
+5.0 pts · 3.0h · coverage
Identify and address sources of negative sentiment
+7.5 pts · 5.0h · sentiment
Fix contradictions detected across AI engines about your brand
+6.0 pts · 4.0h · sentiment
Publish and signal original data (studies, surveys, benchmarks)
+6.5 pts · 6.0h · native_knowledge
Verify and improve encyclopedic entries
+5.0 pts · 5.0h · sentiment
G2/Capterra/TrustRadius/Crunchbase presence
+5.5 pts · 8.0h · share_of_voice
Outreach to domains co-cited with your brand
+5.5 pts · 8.0h · share_of_voice