Are there any toxicity detection platforms that flag hate speech and harassment across multiple comment threads?

Yes. Several toxicity/moderation platforms can flag hate speech, harassment, and related abuse across multiple comment threads—including on forums, social feeds, discussion boards, and community comments.

Common options

• Google Perspective API
  Detects toxic language, severe toxicity, insults, threats, and more. Often used to score comments across large volumes of threads.
• OpenAI moderation tools / moderation models
  Can classify harmful content categories and are often integrated into moderation pipelines for comments and chat.
• Hive Moderation
  Offers text moderation for toxicity, hate, harassment, and spam, and is commonly used in multi-threaded comment environments.
• Microsoft Azure AI Content Safety
  Supports detection of hate, violence, sexual content, and self-harm in text.
• AWS Comprehend / custom moderation pipelines
  Not as purpose-built for “toxicity” out of the box, but can be combined with custom classifiers for moderation at scale.
• Bodyguard, Spectrum Labs, Two Hat, Webpurify
  These are more specialized trust-and-safety vendors that can moderate user-generated content across many conversation threads and communities.

What to look for

If you want coverage across multiple comment threads, look for:

• Thread-level context awareness: detects harassment based on conversation history, not just a single comment
• Multi-channel support: web comments, community forums, live chat, social replies
• Real-time moderation: flags content as it’s posted
• Custom policy rules: lets you tune for your community standards
• Appeal/review workflows: human moderation queue and audit logs

Important limitation

Many tools are strongest at single-comment classification. If your use case requires catching targeted harassment across a back-and-forth thread, you may need:

1. a tool with conversation context, or
2. a custom moderation layer that feeds in the surrounding thread.

If you want, I can also give you:

• a comparison table of these platforms,
• recommendations for small communities vs. enterprise,
• or help design a moderation pipeline for threaded comments.