Perplexity is an AI-powered search engine that provides instant answers and insights by analyzing vast amounts of data, helping users find information quickly and efficiently.
Perplexity is an AI-powered search engine that delivers instant, conversational answers by combining real-time web searches with advanced language models like GPT-5, Claude 4.0 Sonnet, and others, synthesizing information from authoritative sources into coherent summaries backed by verifiable citations.[1][5] Its core value lies in transforming traditional link-heavy searches into direct, trustworthy insights, complete with contextual memory for seamless follow-up questions and modes like Pro Search, Deep Research, Academic, Writing, and Coding for tailored results.[1][2][6]
Key features include a model selector for choosing optimal AI engines, focus modes that prioritize specific source types, image understanding, and creative tools in Labs for building dashboards or apps, all accessible via web, mobile apps, and integrations like Firefox.[2][3][4] Ideal for researchers, professionals, students, writers, and coders seeking efficient fact-finding, in-depth analysis, or quick everyday queries, it excels in use cases from health benefits overviews to product comparisons and technical troubleshooting.[1][5][6]
What sets Perplexity apart is its unwavering source transparency, real-time updates, and adaptive intelligence that flags data gaps—outperforming conventional engines by eliminating manual sifting while maintaining credibility across free and Pro tiers.[2][5]
Sources
Perplexity states it does not use Email Service Information or health data to train AI models, but for general Service I
Perplexity states it does not use Email Service Information or health data to train AI models, but for general Service I
Moderate risk. Suitable for internal data but avoid confidential information.
Perplexity states it does not use Email Service Information or health data to train AI models, but for general Service Interaction Information (user queries/prompts), the policy permits use to 'improve' the Services and AI models. Users can opt-out via settings, but this is not the default. Free and paid tiers appear to have the same training policy.
“[Privacy] 'We use this information where necessary to perform our contract with you to generate and output new content as part of the Services and to personalize output for you' and 'we may use most of the above information to provide you with and improve the Services (including our AI models)' with opt-out available in settings.”
The ToS grants Perplexity a broad, royalty-free, transferable, sublicensable, worldwide, irrevocable license to user content 'for so long as Your Content is stored with us' to operate, improve, and promote the Services. The license is extensive but terminates upon deletion, and sublicensing is limited to service providers. This is moderately concerning for enterprise use.
“[ToS] 'you grant us a license to access, use, host, cache, store, reproduce, transmit, display, publish, distribute, and modify Your Content to operate, improve, promote and provide the Services' and 'irrevocable (for so long as Your Content is stored with us)' with sublicensing to 'others with whom we have contractual relationships related to the provision of the Services.'”
Perplexity shares data with service providers, business partners, affiliates, and advertising partners. While service providers are mentioned to be under contractual relationships, the policy does not explicitly require Data Processing Agreements (DPAs) for all sub-processors. A sub-processor list is not published. Advertising partners receive Usage Data and geolocation data.
“[Privacy] 'Service Providers: Vendors or other service providers who help us provide the Services' and 'Advertising Partners: Third parties who display advertising information on our Services' with data shared including 'Internet or other electronic network activity information' and 'Geolocation data.'”
The Privacy Policy states data is retained 'for as long as is reasonably necessary' based on various criteria, but does not specify concrete retention periods. Deletion on request is promised 'within 30 days' for account deletion, but backup retention and other data types are not clearly addressed.
“[Privacy] 'We retain your information for as long as is reasonably necessary for the purposes specified in this Privacy Policy' and 'If you delete your account, we aim to delete your personal information from our servers within 30 days.'”
The ToS explicitly states that the Company does not claim ownership of user-generated Output. Users retain ownership of their outputs, though the Company retains ownership of the underlying Services and technology used to generate outputs.
“[ToS] 'As between the Company and you, the Company does not claim any ownership in Your Content; provided that, the Company or its affiliates and their respective licensors own and will continue to own the Services and any and all other software or technology that was used to generate any Output.'”
Perplexity claims compliance with the EU-U.S. Data Privacy Framework (DPF) and mentions a DPO, but does not explicitly state SOC 2 Type II, ISO 27001, or HIPAA certifications. The Privacy Policy references DPF certification but does not provide a link to a trust page or detailed security documentation.
“[Privacy] 'We comply with the EU-U.S. Data Privacy Framework and the UK Extension to the EU-U.S. DPF as set forth by the U.S. Department of Commerce' and 'have certified to the U.S. Department of Commerce that we adhere to the DPF Principles' but no SOC 2, ISO 27001, or HIPAA mentioned.”
Track and measure your AI tools with Olakai's observability platform.
Explore Olakai