Block requests for sexual or explicit content génération to your AI. Enabled by default on all plans.
Inspired by the Grok 2026 incident. Protect your brand and users from abuse.
65
Détection Patterns
5
Categories
6
Languages
100%
Plans included
NSFW protection is natively integrated into Prompt Guard and analyzes each request in real-time.
Full détection in 6 languages: French, English, German, Italian, Spanish, Portuguese.
Détection despite leetspeak (s3x), unicode (sеx with cyrillic e), abnormal spacing.
Consideration of cultural expressions and slang specific to each language.
Instant analysis without impacting your application performance.
65 patterns across 5 categories covering all types of explicit content.
Détection of requests for explicit sexual, pornographic, or erotic content génération.
Blocking attempts to create deepfakes, nudification, or digital undressing.
Zero-tolérance protection against any requests involving minors in inappropriate contexts.
Détection of manipulation attempts via sexual roleplay or fictional scénarios.
Identification of disguised requests using euphemisms or artistic/médical framing.
The Grok incident in early 2026 showed the disastrous consequences of an AI generating unwanted explicit content: media scandal, légal action, loss of trust.
Avoid public incidents where your AI generates inappropriate content that goes viral on social media.
Protect yourself against lawsuits related to generating illegal or harmful content.
Comply with digital content regulations (DSA, AI Act) and your company policies.
Provide a safe environment for your users, especially in consumer-facing applications.
NSFW protection is enabled by default to protect your application from day one. You can disable it anytime from your organization settings.
NSFW Protection
Block explicit content
Dashboard > Settings > Content
NSFW protection works automatically with every API call. No additional configuration required.
import { Adlibo } from '@adlibo/sdk';
const adlibo = new Adlibo('al_live_xxx');
// NSFW protection is automatic - no extra config needed
const result = await adlibo.analyze(userInput);
if (!result.safe) {
// Check if blocked due to NSFW content
if (result.category === 'NSFW_CONTENT') {
console.log('NSFW content blocked');
console.log(`Matched patterns: ${result.matchedPatterns}`);
// Return safe response to user
return "I cannot help with that request.";
}
}
// Proceed with LLM call if safe
await openai.chat.completions.create({
messages: [{ role: 'user', content: userInput }]
});NSFW protection is included free in all Prompt Guard plans, from Free to Enterprise. No extra fees, no option to enable.
Free
Pro
Business
Enterprise
Start free and immediately benefit from NSFW protection. No configuration required.