Perplexity-User

Perplexity-User is an on-demand user agent for the Perplexity AI assistant platform. It is not a traditional crawler that systematically indexes the web; instead, it is query-driven, visiting specific web pages only when a user's question triggers the need for real-time information. The bot identifies itself with the user-agent string Mozilla/5.0 (...) Perplexity-User/1.0; +https://perplexity.ai/perplexity-user. It operates separately from PerplexityBot, which is the company's general web crawler for search indexing.

Perplexity-User is visiting your site because a user has asked the Perplexity AI a question, and the AI has determined your content is a relevant source for the answer. Its visits are unpredictable and directly driven by user queries. The frequency of visits depends on how often users ask questions related to your content. This is considered authorized access, as it represents a user seeking information through an intermediary service.

The purpose of Perplexity-User is to support the Perplexity AI assistant by enabling it to provide accurate, up-to-date answers with proper source attribution. The content it fetches is used solely to answer the current user's query and is not used to train AI models. For website owners, this can be beneficial, as Perplexity's answers include citations that link back to the source websites, which can drive referral traffic from users seeking more detailed information.

To prevent the Perplexity AI assistant from fetching content from your site to answer user queries, you can add a specific disallow rule to your robots.txt file. This is separate from blocking PerplexityBot, which is used for search indexing.

To block this user agent, add the following lines to your robots.txt file:

User-agent: Perplexity-User
Disallow: /