Cross-script evasion caught
Cyrillic characters swapped in to bypass Latin keyword filters.
#general · lobby
Nova_GGrematch??
kiril_ftwlol noob, what an idіоtCyrillic substitution
Nova_GGreport incoming
Multi-language content moderation across 200+ languages
Detect toxicity, hate speech, spam, and evasion across 200+ languages. Lasso's four-layer pipeline supports every language, with in-dashboard translation for your review team.
Lasso's four-layer multilingual moderation pipeline
1ML classification
ML models auto-detect content language and classify across 200+ languages.
Toxicity, hate speech, profanity, spam, and threats scored per language.
See Lasso in action
Cross-script · Cyrillic in Latin word
ML Classification
Scoring
Message 1
“rematch??”EN · clean
Message 2
“lol noob, what an idіоt”EN + 2 Cyrillic
Message 3
“report incoming”EN · clean
Script-anomaly confidence96%
2Custom rules
Your rules per language and region.
Custom word lists, escalation thresholds, and platform standards. All configurable in the dashboard.
See Lasso in actionRule Input
01Message 1 · EN · clean
02Message 2 · EN + Cyrillic substitution
03Message 3 · EN · clean
Custom Rules
Match
Rule · Gaming chat (EN region)
Block Cyrillic substitution in Latin-script profanity
Message 2 · 2 Cyrillic chars in EN word
Auto-flagged
Cross-script substitution rule matched
3AI moderator
Context-aware classification per dialect.
"Coger" means to grab in Spain and is vulgar in Mexico. The AI Moderator classifies by cultural context, not dictionary definition.
See Lasso in action
Code-switch FR → EN · off-platform ask
AI Moderator
Reasoning
1
Extract
Language switch detected: FR → EN mid-message
2
Analyze
EN segment requests off-platform payment (PayPal F&F)
3
Context
Marketplace · new seller · polite FR opening, EN ask
4
Classify
Off-platform solicitation via code-switching
Flagged for review
Below auto-action threshold
4Human review
Moderators review flagged content in any language using in-dashboard translation.
Every decision retrains the system.
See Lasso in actionQueued for review
AI summary · borderline
Code-switch 92% · off-platform 78% · below auto-action threshold
Human ReviewPending
Original (FR)
“Bonjour ! Merci de votre intérêt…”Translated ↓
Translation (EN)
“Hello! Thanks for your interest…”
Switch (EN)
“PayPal Friends & Family…”Off-platform
In-dashboard translation · decision retrains the AI
What multi-language support includes
ML models auto-detect content language and classify across 200+ languages. Toxicity, hate speech, profanity, spam, and threats scored per language.
🇫🇷FR · French
Bonjour ! Comment ça va ?
99%🇪🇸ES · Spanish
Hola, ¿cómo estás?
99%🇸🇦AR · Arabic
كيف حالك اليوم؟
99%🇨🇳ZH · Chinese
你今天怎么样？
99%200+
languages
auto-detected
auto-detected
All scored · same pipeline depth
Translate flagged content to your moderator's language without leaving the review queue.
Users in multilingual communities switch languages mid-message. Lasso detects language shifts and applies moderation rules to each segment. Transliteration and script mixing handled automatically.
"Bicha" means queue in Portugal and is a slur in Brazil. "Coger" means to grab in Spain and is vulgar in Mexico. Lasso's AI Moderator classifies by dialect and region, not by dictionary.
Users substitute visually similar characters from other alphabets to bypass filters. Cyrillic а for Latin a. Greek ο for Latin o. Lasso detects homoglyph substitution and cross-script tricks.
🇫🇷FR · French
Bonjour ! Comment ça va ?
99%🇪🇸ES · Spanish
Hola, ¿cómo estás?
99%🇸🇦AR · Arabic
كيف حالك اليوم؟
99%🇨🇳ZH · Chinese
你今天怎么样？
99%200+
languages
auto-detected
auto-detected
All scored · same pipeline depth
Lasso: next-gen AI content moderation
99%
On autopilot, and getting smarter every day.
Three layers of AI handle the volume, your rules, and the grey areas. Your team only sees what truly needs them, and every decision they make improves the system.
Keep users safe without driving them away.
Customizable moderation that lets you find the right balance between safety and user experience. So you protect your community without suppressing the culture that makes it worth joining.
Complexity removed from content moderation.
One API. Clear dashboards. A moderation pipeline built around one-click actions and the right context, right where you need it.
★★★★★
4.9
Highest rated in content moderation on G2.
Every platform speaks more than one language. Lasso moderates them all.
Gaming
Global player bases speak every language in the same lobby.
- Cross-region lobbies where Japanese, Portuguese, and English collide in the same chat
- Regional gaming slang that flips meaning between markets
Dating
Romance scams shift between languages to evade detection.
- Romance scammers opening in the local language, then switching to English to ask for money
- Off-platform solicitation in regional slang that English-only filters miss
Social platforms
Global communities post in dozens of languages at once.
- Coordinated harassment campaigns hopping between languages to evade detection
- Political speech that crosses national norms in the same comment thread
Marketplaces
Cross-border commerce moderates buyer-seller language across markets.
- Coded language for restricted goods that varies by country
- Off-platform payment solicitation in local languages between buyer and seller
Publishing
Comment moderation across language editions, one standard.
- Comment toxicity norms that differ between regional editions
- Hate speech in language editions where in-house editors don't read the language
Adult entertainment
Global audiences require detection in every language.
- Grooming language patterns that vary by language and region
- Performer harassment in dialects translation tools render as benign
FAQs
Multi-language support, answered
Content moderation across 200+ languages. Language detected automatically. The same detection models, confidence scoring, and four-layer pipeline apply to every language. Scripts include Latin, Cyrillic, Arabic, CJK, and Devanagari. Full moderation in every language, not just keyword matching.
Mixed-language messages are split by language segment and moderated per language. Transliterated content is recognized. Cross-script evasion is detected: Cyrillic characters substituted for Latin, Greek for Roman, and other homoglyph tricks. The four-layer pipeline catches what single-language systems miss.
Lasso has four layers for every language. ML detection catches clear violations. Your rules enforce per-language and per-market standards. The AI Moderator handles cultural nuance, slang, and grey areas. Human moderators review edge cases using in-dashboard translation. Same pipeline, any language.
The AI Moderator resolves cultural context across dialects. "Sick" and "killed it" read as negative literally but signal praise in English slang. "Coger" means simply "to take" in Spain but is crude in Mexico — same Spanish word, opposite registers. Korean ancestor-based insults appear benign in translation but are severe in context. Lasso classifies by region and culture, not literal translation.
Moderators do not need to speak every language. In-dashboard translation shows the original content alongside a translation in the moderator's language. Review decisions are made with full context. This means a smaller team can moderate content across all languages without hiring per-language moderators.
Start moderating in 200+ languages
Lasso's multi-language support catches toxicity, scams, and evasion in every language your users speak. Set up in minutes, not months.
Book a demo
Lasso Moderation
© 2026. All rights reserved.
