Email Safety Advisor
durch iVerifi Limited
AI phishing protection for Outlook. Warns staff before they interact with suspicious emails.
Email Safety Advisor analyses every email the
moment it is opened in Microsoft Outlook and
shows staff exactly why it is suspicious —
before they click anything.
HOW IT WORKS
When a user opens an email, the add-in
automatically assesses it using Azure OpenAI
GPT-4o in under 2 seconds. It checks sender
authentication (DMARC, SPF, DKIM), Microsoft
365 security headers, domain age, urgency
language patterns, and display name spoofing.
FOUR PROTECTION MODES
Mode 0 — Silent (score 0-25)
Safe email. Zero interruption. Logged only.
Mode 1 — Passive badge (score 26-50)
Small indicator in toolbar. Click to expand.
Mode 2 — Amber warning (score 51-75)
Dismissible banner before reading further.
Mode 3 — Full AI panel (score 76-100)
Complete advisory appears automatically
before any user interaction.
WHAT MAKES IT DIFFERENT
Approximately 70% of emails receive Mode 0 —
completely silent. Warnings carry real weight
because they are rare and specific to that
exact email. No alert fatigue.
One-click Report to SOC sends a complete
technical incident record to your security
team automatically.
All assessments including silent ones are
logged for compliance and audit reporting.
DEPLOYMENT
Deploys in 30 minutes via Microsoft 365
Admin Center. No endpoint agents. No VPN
changes. No user installation required.
DATA PRIVACY
Minimum necessary data — sender identity,
subject, and 500-character excerpt only.
Personal names and financial details are
automatically tokenised before AI analysis.
Full message body and attachments are never
accessed. Azure UK South data residency.
UK GDPR compliant.
This add-in requires an active iVerifi subscription
to use the AI classification service. A 12-week pilot
is available for £2,500. Monthly subscriptions start
at £3 per user per month. Visit safety.iverifi.io
for full pricing details.
Patent Pending — UK Application GB2610008.1
iVerifi Limited — Registered in England and Wales
Appfunktionen
- Es kann Daten über das Internet senden.
- Diese App kann auf persönliche Informationen in der aktiven Nachricht zugreifen und diese ändern, beispielsweise Text, Betreff, Absender, Empfänger und Anlageninformationen. Andere Elemente in Ihrem Postfach können weder gelesen noch geändert werden. Weitere Informationen zur Verwaltung und Verarbeitung persönlicher Informationen finden Sie in der Datenschutzrichtlinie des Entwicklers.