Tisane

About Tisane

Tisane is an automatic content moderation API for detecting problematic user-generated content, including cyberbullying, hate speech, criminal activity, allegations, profanities, sexual harassment, and adult-only references, across 35 languages. It includes entity extraction, topic modelling, language identification, and built-in automatic translation, and is designed to counter algospeak, character substitution, and other adversarial text manipulations that evade rule-based filters. Output includes explanations for flagged content, and the service is described as compliant with DSA standards.

The product is aimed at trust and safety teams moderating user-generated content in contexts such as gaming, online marketplaces, threat intelligence, and online communities. It is available as SaaS, on-premises deployment, or embedded in a server, desktop, or mobile application, and offers ready-made integrations including Slack, Webex, Minecraft, Discord, YouTube, Maltego, and Zapier. Tisane Labs also offers consulting projects for custom multilingual moderation challenges.

Key features

• Detects abusive content such as hate speech, cyberbullying, harassment, and criminal activity in text
• Explainable moderation results with severity ratings and reasoning for each flag
• Counters algospeak, character substitution, and adversarial text manipulation
• Built-in automatic translation to reveal obfuscated profanity and slurs across languages
• Entity extraction and granular topic modelling
• Supports 35 languages with ready-made integrations for Slack, Webex, Maltego, Minecraft, Zapier, and more

Frequently asked questions