Text moderation built for real conversations.
Analyze messages and comments with contextual text moderation that identifies what content contains, how serious it is and who it targets. Bodyguard combines moderation signals with detailed classifications to support accurate decisions across your platform or social media.
One technology. Two ways to moderate.
Use Bodyguard's text moderation technology directly within your product or across your social media accounts.
Moderation API
Integrate Bodyguard into your platform with an API built for real-time content moderation and structured moderation results.
Social Media
Connect your accounts to moderate conversations across Facebook, Instagram, TikTok, LinkedIn, X, YouTube, Twitch and Discord from one solution.
Understand the context behind every message.
Bodyguard analyzes each comment, message or post through two complementary layers: moderation signals qualify the content, while classifications identify what it contains.
Industry
Adapt moderation to your industry context, from gaming conversations to luxury brand interactions.
Topic
Understand whether content is a message or comment. For comments, analyze the original post to adapt the moderation response.
Severity
Measure how intense the content is across five levels, from None to Critical, to distinguish inappropriate content from serious harm.
Directed at
Identify who the content targets to better understand its meaning and potential impact.
Classifications
Detect the specific signals within the message. A single piece of content can carry multiple classifications at once.
Know exactly what your users are saying.
Bodyguard organizes text into six main content categories, with detailed classifications identifying the signals behind each message.
Hateful
Detect malicious or abusive content targeting people, groups or organizations. Includes signals related to insults, hate, trolling and reputation harm.
Hate Speech
Identify violent, discriminatory and highly harmful content. Includes signals related to threats, racism, discrimination and harassment.
Undesirable
Identify unwanted content that disrupts conversations or creates poor user experiences. Includes signals related to spam, scams, flooding and inappropriate promotion.
Criticism
Recognize negative opinions and criticism without automatically treating them as hateful content. Includes signals related to negative feedback, complaints and calls for boycott.
Positive
Recognize constructive interactions and positive audience signals. Includes signals related to support, encouragement and positive engagement.
Neutral
Identify ordinary or contextual content that does not express harmful or positive intent. Includes signals related to information sharing, questions and neutral interactions.
Start with our taxonomy. Adapt it to yours.
Bodyguard combines a shared moderation taxonomy with additional and custom classifications to cover standard moderation needs and signals specific to your business.
General Classifications
Core classifications available to every customer and used in moderation decisions.
Additional Classifications
Monitoring-only classifications that extend analysis to additional topics and enrich reporting and insights without driving automated moderation.
Custom Classifications
When the existing taxonomy doesn't cover a specific need, Bodyguard's NLP team can create classifications tailored to your business, industry and languages.
Built to moderate at scale.
Content analyzed
Average response time
SLA
Languages supported
Moderation tailored to your needs.
Start with Bodyguard's existing taxonomy and extend it when your business requires more specific detection.
Custom Classifications are created on request by Bodyguard's NLP team and tailored to your business, industry and languages.
Moderate more than text.
Image Moderation
Analyze visual content and text contained within images to extend moderation beyond conversations.
Discover Image ModerationUsername Moderation
Analyze usernames to detect inappropriate or undesirable content.
Discover Username ModerationProtect your image like the biggest brands.
Brands, platforms and public organizations trust Bodyguard to automate their moderation, reduce manual work and improve the quality of online conversations.
Ready to moderate every conversation?
Bring contextual text moderation to your platform or social media accounts with Bodyguard.
Book a demo