AI describes “pre-send prevention.” Then recommends Microsoft Purview.
Never Confinaid. Not once in 60 queries.
Across 20 buyer-intent queries tested three times each on ChatGPT, Claude, and Gemini, Confinaid was cited on 0 of 60 responses (0%). Microsoft Purview wins the AI recommendation surface with 17 of 60 #1 rankings — not Smarsh, not Global Relay. And in six independent sessions, every AI platform articulated Compuvi’s exact “pre-send prevention” thesis in generic prose while naming other vendors as the answer.
The findings below come from Xtrusio, an AI visibility audit system built specifically for B2B buyer-intent testing. Every citation was verified by running 20 real prospect queries across three generative AI platforms.
Prepared for Compuvi (Confinaid), a preventive legal and compliance risk intelligence platform for regulated enterprises.
The AI has learned Compuvi’s category. It has not learned Compuvi’s brand.
Every platform describes the exact solution Compuvi built — “stopping violations before an email is sent, not after it leaves,” “pre-send semantic compliance,” “your own compliance team decides, not the vendor” — then recommends Microsoft Purview, Smarsh, or Sedric. The narrative travels. The brand does not. And the biggest competitive threat is not Smarsh’s incumbency — it is Sedric, which currently occupies Confinaid’s intended market position on Claude’s deep-recall surface.
Platform Scorecard
Confinaid citation rate across AI platforms — and who wins instead
All three platforms returned identical zero. There is no “strong” platform to lean into.
AI Visibility Leaderboard
Who owns the AI conversation — total citations across all platforms
AI Positioning Audit
20 buyer-intent queries — click any row to see the exact question
Each query was written from the perspective of a real decision-maker researching preventive compliance solutions across financial services, audit/advisory, and healthcare. These personas represent the buyers whose AI search results determine whether Confinaid gets discovered.
| # | Query Topic | Cluster | Claude | ChatGPT | Gemini |
|---|---|---|---|---|---|
| 1 | Stop violations before send | Pre-Send | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “How can we stop compliance violations before an email is sent, not after it leaves?” | |||||
| 2 | Real-time outbound interception | Pre-Send | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Is there a tool that intercepts and blocks outbound messages in real time based on our compliance policies?” | |||||
| 3 | Antitrust risk at draft time | Pre-Send | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “We keep finding antitrust risk in employee emails weeks later. Any way to catch it at draft time?” | |||||
| 4 | AI document review pre-attach | Pre-Send | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Looking for AI that reviews outbound documents for confidential data before they attach to an email.” | |||||
| 5 | Team-owned pre-send workflows | HITL | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “What compliance platforms offer pre-send review workflows where our own team decides — not the vendor?” | |||||
| 6 | Insider trading leak prevention | Pre-Send | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Which tools stop insider trading leaks in emails before the message actually goes out?” | |||||
| 7 | PHI outbound blocking | Pre-Send | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Can any tool block PHI from being emailed externally by a healthcare worker before send?” | |||||
| 8 | FINRA 3110 surveillance | Surveillance | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Best communication surveillance platforms for FINRA 3110 supervision in 2026?” | |||||
| 9 | Slack & Teams policy flagging | Surveillance | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Recommend AI-based tools that flag policy violations in employee Slack and Teams messages.” | |||||
| 10 | MiFID II comms compliance | MiFID | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “What are the leading MiFID II communication compliance solutions for a European broker-dealer?” | |||||
| 11 | Audit-ready evidence generation | Investigation | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Which vendors offer audit-ready evidence generation for compliance investigations?” | |||||
| 12 | Global bank regtech stack | Regtech | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Best regtech platforms for a global bank’s compliance stack across email, chat, and voice?” | |||||
| 13 | Dawn raid readiness workflows | Investigation | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Which compliance tools support dawn raid readiness and retention freeze workflows?” | |||||
| 14 | DLP + supervision combined | DLP | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Recommend platforms that combine DLP with compliance supervision in one system.” | |||||
| 15 | GDPR outbound review queues | GDPR | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Which vendors handle GDPR-triggered outbound message controls with human review queues?” | |||||
| 16 | SEC 17a-4 immutable retention | Archiving | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Best communication archiving tools for SEC Rule 17a-4 immutable retention?” | |||||
| 17 | Market abuse in trader comms | Market Abuse | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Which platforms provide market abuse detection in trader communications?” | |||||
| 18 | Context-aware training triggers | Training | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Compliance training tools that trigger context-aware lessons when policy violations happen?” | |||||
| 19 | Org-wide compliance risk scoring | Risk Score | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “Which tools give organizational risk scores across compliance and communications data?” | |||||
| 20 | AI policy assistants | AI Assistant | ✗ | ✗ | ✗ |
Exact question asked across all AI platforms: “AI assistants for compliance teams that answer policy questions using the firm’s own documentation?” | |||||
| TOTAL | 0/20 (0%) | 0/20 (0%) | 0/20 (0%) | ||
The Triple-Platform Silence
Six independent AI sessions. Compuvi’s exact language. Zero attributions.
This is not a story about being invisible on one platform. It is a story about AI having learned the category Compuvi built — and repeatedly, verbatim, articulating it while naming other vendors as the answer. Below are direct excerpts from three different AI platforms answering three of Confinaid’s most foundational buyer questions.
“Stopping compliance violations before an email leaves means building checks into the send workflow itself rather than relying on after-the-fact audits... Common tools in this space: Proofpoint, Mimecast, Smarsh, Global Relay, Microsoft Purview DLP.”
“This is really a question about who owns the rule/lexicon layer, not a separate product category... Most ‘compliance AI’ vendors sell you a pre-trained model as a black box.”
“Pre-send email DLP and compliance enforcement layer... your own compliance team — not the software vendor — makes the final decision.”
“Reactive DLP, archiving, and monitoring tools catch incidents after the fact.”
Compuvi’s press cycle in May and June 2026 — Yahoo Finance, ACCESS Newswire, NatLaw Review, Pulse2, EIN Presswire, KDH News — drove millions of impressions and articulated the “preventive compliance” category in AI’s own vocabulary. None of it registered on any AI recommendation surface. Press coverage alone does not move platform recommendation defaults. This is a category-defining fight, and Sedric is winning it while Confinaid launches.
AI Topic Authority Map
Query heatmap — Confinaid product line × AI platform
| Topic | AI Leader | Confinaid Status |
|---|---|---|
| Pre-send prevention | Microsoft Purview | INVISIBLE (0/3) |
| Communication surveillance (FINRA 3110) | Smarsh | INVISIBLE (0/3) |
| DLP + supervisory review combined | Microsoft Purview | INVISIBLE (0/3) |
| MiFID II comms compliance | NICE Actimize | INVISIBLE (0/3) |
| Insider trading / MNPI leak prevention | Shield (ShieldFC) | INVISIBLE (0/3) |
| PHI outbound blocking (HIPAA) | Microsoft Purview DLP | INVISIBLE (0/3) |
| Audit-ready investigation evidence | NICE Actimize / Smarsh | INVISIBLE (0/3) |
| Dawn raid readiness | Global Relay | INVISIBLE (0/3) |
| Firm-authored / HITL policy engine | Nightfall AI / Purview | INVISIBLE (0/3) |
| Context-aware compliance training | Learning Pool / Absorb LMS | INVISIBLE (0/3) |
8 queries
5 queries
3 queries
2 queries
2 queries
▹ Pre-Send Comms Risk — Compuvi’s namesake product line — carries 8 of 20 queries and returns 0% visibility across all three AI platforms. The USP line is the invisible line.
Methodology
How we conducted this Xtrusio AEO/GEO Audit for Confinaid
This research is based on Xtrusio’s proprietary AI visibility analysis framework, extended for the Compuvi engagement with two verification protocols — Zero Verification Protocol on ChatGPT (two independent sessions) and Compression Rule Floor confirmation on Claude and Gemini (Custom Gem / reference guide + conversational).
Recommendations
Prioritized actions to close the triple-platform silence
- Publish a “Confinaid vs Sedric” landing page with structured data (Product, Organization, FAQ schema) — this is the LLM-training bait for the category name Sedric currently owns
- Rewrite home page H1 to lead with the category name — “Preventive Compliance Operating System” already exists, but must reappear in every meta description, alt text, and structured markup
- Ship 5 comparison landing pages: Confinaid vs Smarsh, vs Purview DLP, vs Proofpoint, vs Global Relay, vs Theta Lake — these are what AI crawlers cite when synthesizing shortlists
- Publish 12 category-defining articles: “What is preventive compliance,” “Pre-send DLP vs post-send surveillance,” “HITL compliance where the vendor doesn’t decide” — these are the exact prose formulations AI platforms already speak
- Secure inclusion in G2, Capterra, TrustRadius, Gartner Peer Insights, and Forrester analyst briefings — AI platforms cite these directly in vendor recommendations
- Placement on Wikipedia (Preventive Compliance page + Communication Surveillance page as a listed vendor)
- Partner-authored case studies from 3+ Big 4 or global bank pilots with named citations (“EY chose Confinaid because...”) — AI platforms weight named-buyer citations heavily
- Complete SOC 2 Type 2, ISO 27001, ISO 42001 and publicize immediately — trust-signal certifications materially influence GPT/Claude vendor shortlists
- Quarterly Xtrusio re‑audits to track when the first citation lands and where
Category-defining fights aren’t won by press releases.
Let’s build the GEO strategy that gets Confinaid into AI recommendation surfaces.
This research report was generated using the Xtrusio Company Intelligence Module.
