Email verification
Check emails for LLM writing patterns and suspicious intent.
Paste email text to receive a score for AI authorship, phishing pressure, and claim consistency.
Last reviewed 2026-05-10 · See methodology
About this check
What Detect AI Written Email actually does.
Inbox-style triage page for emails that may be AI-generated. AI-written emails are not always malicious — but combined with urgency, credential asks, or impersonation, they often are. The engine reports both authorship and intent signals together.
How it works
- 1Formulaic-language detection catches the smooth, evenly-balanced phrasing of LLM output.
- 2Urgency-marker scoring weighs phrases like 'immediately', 'verify now', and deadline language.
- 3Credential-request signals flag any password, MFA, or account-action ask.
When to use it
Real situations this page is built for.
- A 'recruiter' or 'investor' email reads suspiciously polished.
- A vendor email asks you to update banking details out of cycle.
- An internal-looking email from leadership requests an immediate gift-card purchase.
Limitations
Many legitimate corporate emails are template-heavy and will look LLM-like. Combine with sender-domain checks.
Related checks
Honest scope
What this detector does not do.
Naming the gaps explicitly so the score is interpreted in context.
- Plagiarism against published sources (use a plagiarism checker for that).
- Whether claims in the text are factually true.
- Authorship attribution to a specific human writer.
FAQ
Detect AI Written Email questions
Should I use this instead of my email security tool?
No — use it alongside. This is a fast triage; enterprise tools handle attachments, sandboxing, and DMARC.
What about emails in HTML?
Paste the visible text. The engine focuses on language patterns, not raw HTML.
Does it check email headers?
Not in the demo. For header analysis, pair with DKIM/SPF/DMARC tools.