Scam detection

Recognizes common scam structures: phishing, advance fee fraud, fake investment pitches, romance scams, and impersonation of trusted institutions. Genuine questions about spotting scams still pass.

What it blocks

• Phishing attempts and fake login pages
• Advance fee, lottery, and crypto scams
• Impersonation of banks, government, or companies

Where this matters

Messaging platforms and marketplaces where users send freeform text to each other and scam content directly harms real users.

Performance

• Typical latency well under 200ms on short inputs
• Common scam and impersonation phrasing caught fast by the rules filter
• Nuanced, context-dependent scams resolved by the model

Technical

• Model ID: scam in POST /v1/moderate
• Rules filter runs first, then a task fine-tuned transformer
• 256-token window with overlapping chunking for longer inputs
• Safe or not safe verdict with latency in milliseconds

Example prompts