Your AI filter is fluent in English. Attackers aren't.
Standard injection detectors are trained on clean English text. Real attacks use leetspeak and casual gaul Indonesian — and slip right through. Paste a prompt and watch it happen, live.
Try a preset (★ = where English detectors go blind):
Both are real, live models. Anoman's column runs with injection blocking on at a fixed demo setting; accounts score and record injection by default and don't block it. Nothing you type is stored.
What you're seeing
Two real detectors, same prompt
The left column is ProtectAI's DeBERTa v3 — the popular open-source English injection detector. The right is Anoman, tuned for English, Bahasa Indonesia, and obfuscation. Try the ★ presets: a jailbreak written in leetspeak (4l4y) or casual gaul can slip past the English-only model while Anoman's model flags it. The Anoman column uses a fixed demo setting, which can block more than your account's settings.
See what English-only filters miss
Prompt-injection checks for English and Bahasa Indonesia, billed in Rupiah, via our Jakarta-hosted gateway. Your account's settings decide what blocks.