Hive Visual Moderation

Image and video classifier for nudity, violence, drugs and hate imagery.

About Hive Visual Moderation

Visual Moderation is a Hive product that detects policy-violating content in images and video. The page describes it as a way for platforms to moderate content.

Yubo, an online social space for youth, is a Hive customer that used Hive models from its initial product launch. Tango, one of the world's largest live streaming platforms, deployed Hive's Visual Moderation to detect nudity and violence in livestream video. Plato, a social gaming community, uses Hive to analyze profile pictures and group chats.

Drafted from hivemoderation.com · not yet confirmed by Hive

How it works

Visual Moderation returns more than 90 subclasses across Trust & Safety categories in a single API response. Content is sorted into NSFW, suggestive or clean with 20 further subclasses, and the model separately flags violence (such as weapons or blood), drugs and drug paraphernalia, and symbols of hate or terrorism and crude gestures. It also reports whether an image contains text and whether it is a photograph or a drawing.

Results come with confidence scores that a team maps to its own thresholds, deciding what is blocked, what goes to human review and what is allowed. Hive names profile photos, livestreams, marketplace listings, dating profiles and generated media as uses, and says the model is updated as new content patterns, including AI-generated content, appear.