Perplexity-User

What is Perplexity-User?

Perplexity-User is a specialized user agent operated by Perplexity that supports real-time information retrieval during user interactions with their AI assistant platform. Unlike traditional web crawlers that systematically index the entire web, Perplexity-User is query-driven, visiting specific web pages only when triggered by user questions. It's designed to fetch up-to-date content that helps Perplexity provide accurate, source-cited responses to user queries.

When visiting websites, Perplexity-User identifies itself with the user-agent string: Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; Perplexity-User/1.0; +https://perplexity.ai/perplexity-user). This transparent identification includes a link to Perplexity's documentation about the agent, allowing site owners to verify its legitimacy.

Perplexity-User operates distinctly from PerplexityBot (Perplexity's general web crawler). While PerplexityBot performs scheduled crawling for broad indexing purposes, Perplexity-User responds directly to user queries, retrieving information within seconds to minutes to ensure freshness and relevance of the content provided in answers.

Why is Perplexity-User crawling my site?

If you're seeing Perplexity-User in your logs, it means a user has asked Perplexity a question that's relevant to content on your website. The agent visits your site to retrieve specific information that will help answer that user's query accurately. Unlike scheduled crawlers, Perplexity-User's visits are unpredictable and entirely driven by user questions.

The frequency of visits depends on how often users ask questions relevant to your content. Sites with authoritative, factual information on popular topics may see more frequent visits. Perplexity-User typically looks for text content that can provide factual information, explanations, or data points relevant to user queries.

These visits are considered authorized as they represent a user seeking information through an intermediary service, similar to how a person might use a browser to access your content.

What is the purpose of Perplexity-User?

Perplexity-User supports Perplexity's AI assistant by enabling it to provide accurate, up-to-date answers with proper source attribution. When users ask questions, Perplexity-User retrieves relevant content from across the web, allowing the AI to synthesize information from multiple sources and generate comprehensive responses.

The content fetched by Perplexity-User is used solely to answer the current user query and is not used to train AI foundation models. This ephemeral approach means the information is processed for the immediate response and then discarded, rather than being permanently stored or indexed.

For website owners, Perplexity-User can drive visibility and traffic through citations. When your content is referenced in Perplexity's answers, users receive links back to your site, potentially increasing referral traffic from users seeking more information on the topic.

How do I block Perplexity-User?

While Perplexity-User generally ignores robots.txt directives (since it operates on behalf of users making specific requests), you can still implement controls if desired. This behavior is similar to how web browsers typically ignore robots.txt when users directly request pages.

If you wish to manage how Perplexity-User accesses your site, you can implement server-side controls based on the user agent string or IP addresses. Perplexity publishes its IP ranges at https://www.perplexity.com/perplexity-user.json, which you can use to configure access controls in your web server or firewall.

For traditional web crawlers like PerplexityBot (their general indexing crawler), robots.txt directives are respected, and you could use:

User-agent: PerplexityBot
Disallow: /private/

However, since Perplexity-User acts on behalf of human users, blocking it may reduce your content's visibility in AI-powered search results and reduce potential referral traffic from users who discover your site through Perplexity's answers.

If you have concerns about how your content is being accessed or used, Perplexity provides documentation about their data usage policies, emphasizing that content retrieved by Perplexity-User is not used to train AI models.