Perplexity AI

Perplexity AI is an artificial intelligence company best known for its AI-powered answer engine, which searches the web, synthesizes information from multiple sources, and returns conversational answers with citations. Since its launch in 2022, Perplexity has expanded beyond search into research software, AI agents, browsers, enterprise knowledge tools, desktop automation, and developer infrastructure.

The company's broader direction is increasingly clear: Perplexity is trying to move users from finding information to researching it and ultimately acting on it through a single AI platform.

Last updated: September 1, 2026

About Perplexity AI

Perplexity AI is a privately held artificial intelligence company headquartered in San Francisco. It was founded in 2022 by Aravind Srinivas, Denis Yarats, Johnny Ho, and Andy Konwinski. Srinivas serves as CEO, while Yarats is the company's CTO and Ho serves as chief strategy officer. Forbes reported in 2026 that Konwinski no longer holds an active operating position at Perplexity.

The company originally became known for its AI answer engine: users ask questions in natural language, Perplexity searches for relevant information, synthesizes an answer using AI models, and provides citations that allow readers to inspect the underlying sources.

That approach placed Perplexity between a traditional search engine and a generative AI assistant. Traditional search is mainly designed to help users find webpages. Perplexity attempts to perform part of the research and synthesis before presenting the result.

Its product range has since broadened considerably. Perplexity now operates across AI search, advanced research, browser-based AI, agentic workflows, enterprise knowledge tools, desktop automation, and APIs for developers. Its current subscription structure spans free individual use through Pro, Max, Education, Enterprise, and API products.

The company's defining characteristic is therefore no longer simply “AI search.” Perplexity is increasingly building an interface through which users can search, research, create, and execute work.

What Does Perplexity AI Do?

Perplexity's core activity is helping people find, understand, organize, and increasingly act on information.

At the simplest level, a user can ask a question and receive a synthesized answer supported by citations. More advanced products can conduct multi-stage research, create reports, interact with applications, browse websites, work with files, and execute parts of a workflow.

Its main areas of activity can be grouped into five broad categories:

AI Search and Answer Generation

Perplexity's original product searches available information sources and generates a direct response instead of requiring the user to manually open and compare every search result.

Citations remain central to this model because they give users a route back to the source material.

Advanced Research

Perplexity Research is designed for questions that require significantly more investigation than a normal search.

According to Perplexity, Research can perform dozens of searches, examine large numbers of sources, refine its research plan as it works, and synthesize the results into a longer report.

AI Agents and Workflow Execution

Perplexity Computer moves the platform beyond information retrieval.

Rather than only producing an answer, Computer is designed to complete multi-step tasks by combining research with tools, application connectors, file generation, coding, and other actions.

AI-Native Browsing and Desktop Work

Comet integrates Perplexity's AI capabilities into a Chromium-based browser, while Personal Computer extends agent functionality to local files and applications on supported computers.

Perplexity has also been developing local-first agent technology. In August 2026, it announced Portable Computer, designed to run Perplexity Computer on-device and use remote capabilities when necessary.

Enterprise and Developer Infrastructure

Perplexity sells enterprise products for organizations and provides APIs for developers who want to embed web-grounded AI, search, and agent capabilities into other software.

This means Perplexity operates both as a consumer-facing AI service and as infrastructure for businesses and developers.

Perplexity AI Business Model

Perplexity operates a freemium, subscription, enterprise, and API business model built around the same underlying AI-search and research platform.

The free product helps the company attract a broad base of users. Paid consumer plans increase access to advanced models, Research, file tools, agent features, and higher usage allowances.

Enterprise products adapt the same platform for organizations by adding features such as administrative controls, organizational knowledge repositories, collaboration, permissions, and stronger data-management policies.

Developers represent another layer of the business. Rather than requiring users to visit Perplexity directly, the API Platform allows other applications to incorporate Perplexity's search and AI capabilities into their own products.

The model therefore works across several levels:

The strategic advantage of this structure is that Perplexity does not need to monetize only consumer searches. The same underlying capabilities can generate revenue from individual professionals, enterprise teams, software developers, and advanced AI-agent workloads.

How Does Perplexity AI Make Money?

Perplexity generates revenue primarily through paid subscriptions, enterprise accounts, and developer API usage.

Unlike a mature public company, Perplexity does not publish the kind of detailed segment revenue disclosures found in quarterly or annual public-company filings. A precise audited breakdown of revenue by product is therefore not publicly available.

Consumer Subscriptions

Perplexity offers paid plans for users who need more than the free Standard service.

Its current individual plan structure includes:

• Standard

• Perplexity Pro

• Perplexity Max

• Education Pro

Perplexity's August 2026 plan documentation lists Education Pro at $10 per month for verified users and describes Pro and Max as progressively higher-access plans for research, advanced models, file creation, and other AI tools.

Enterprise Subscriptions

Organizations can purchase Enterprise Pro or Enterprise Max licenses.

Enterprise pricing is seat-based. Perplexity's July 2026 pricing documentation lists Enterprise Pro at $40 per month or $400 annually per seat, while Enterprise Max is listed at $325 per month or $3,250 annually per seat.

Enterprise subscriptions also add organizational features such as administration, internal knowledge search, shared resources, role-based controls, and dedicated support.

API Revenue

Perplexity also charges developers to use its API Platform.

This allows companies to incorporate search-grounded AI and agent functionality into their own applications without building the entire search, retrieval, model-orchestration, and citation infrastructure themselves.

API access is separate from ordinary consumer subscriptions.

Advanced Agent Usage

Perplexity's move toward Computer and other agentic products creates another monetization layer because complex, multi-step tasks can consume more computing resources than ordinary search.

This gives Perplexity an economic model that extends from low-cost information retrieval toward higher-value AI execution.

Products and Services

Perplexity has developed a growing portfolio of consumer, professional, enterprise, and developer products.

Perplexity Search

Perplexity Search is the foundation of the platform.

Users ask questions in natural language, and Perplexity retrieves information, generates a synthesized response, and displays citations alongside the answer.

Its value proposition is speed and synthesis: users can often understand the main points of a subject without manually opening every search result first.

Perplexity Research

Research is Perplexity's advanced investigation tool.

It is designed for complex topics where one search is not enough. The system can iteratively search, read documents, reason about what it finds, refine its approach, and produce a structured report.

This makes Research relevant to professional users working on market analysis, competitive research, technology research, current affairs, finance, and other evidence-heavy topics.

Perplexity Computer

Perplexity Computer is the company's main agentic product.

Perplexity describes Computer as a digital worker that can use information and tools to complete tasks rather than merely answer questions. Documented capabilities include research, application creation, email-related workflows, coding, asset creation, and scheduled work.

The conceptual difference is straightforward:

Search finds and explains information.

Research investigates complex questions.

Computer attempts to complete work.

Comet Browser

Comet is Perplexity's Chromium-based AI browser.

It combines normal web browsing with Perplexity-powered AI functionality and supports familiar browser features such as bookmarks and most Chrome extensions.

Its strategic importance is larger than the browser itself. Comet places Perplexity directly inside the browsing environment instead of requiring users to visit a separate AI-search website.

Personal Computer and Portable Computer

Personal Computer extends Perplexity Computer to local files and applications.

Perplexity's documentation describes the web-based Computer product as cloud-based, while Personal Computer adds access to a user's local machine, including files, applications, and desktop context when permissions are granted.

The company has also introduced Portable Computer, a local-first initiative intended to run agent workloads on-device and call remote resources when necessary.

These products show that Perplexity is expanding beyond web search into operating-system and device-level AI interaction.

Perplexity Enterprise

Enterprise Pro and Enterprise Max adapt Perplexity for organizations.

They add capabilities such as organizational file repositories, collaboration tools, seat management, administrative controls, permissions, and expanded access to Research and file/app creation. Perplexity also states that Enterprise customer data is not used for model training.

Perplexity API Platform

The API Platform allows developers to integrate Perplexity capabilities into other applications.

This gives the company a role not only as a consumer AI service but also as a provider of AI infrastructure.

Customers and Markets

Perplexity serves several distinct customer groups.

Individual Consumers

The free Standard product targets general users who want conversational answers, current information, and citations without paying for a subscription.

This places Perplexity directly in the consumer search and AI-assistant markets.

Professionals and Knowledge Workers

Pro, Max, Research, and Computer are more relevant to people who use AI regularly for research, analysis, writing, development, or complex knowledge work.

Perplexity has historically highlighted academics, students, and knowledge workers as important users of its research products.

Students and Educators

Education Pro targets verified students and educators.

It includes Pro functionality along with education-focused features such as Learn Mode.

Enterprise Customers

Enterprise products target organizations that want AI research capabilities alongside internal company knowledge, role-based access, administration, and collaboration.

Perplexity's Enterprise platform supports organization-level roles and permissions, allowing administrators to control which users or groups can access particular features.

Developers

The API Platform targets software companies and developers that want current web information and AI-search functionality inside their own products.

Perplexity therefore participates in both the end-user AI market and the developer infrastructure market.

Competitors

Perplexity competes across several overlapping markets rather than against one single category of company.

The browser market illustrates how these categories are converging. Reuters reported that OpenAI's Atlas entered a growing field of AI-enhanced browsers that already included Perplexity's Comet, Brave, and Opera Neon, while Google has continued integrating Gemini into Chrome and Search.

Competition is therefore not simply “Perplexity versus Google Search.” Perplexity is competing for multiple parts of the user's digital workflow: search, research, browsing, AI assistance, enterprise knowledge, and agent execution.

Company History

Perplexity's history is short compared with established technology companies, but its product direction has changed quickly.

Reuters reported in September 2025 that Perplexity had secured investor commitments for $200 million at a reported $20 billion valuation, based on reporting by The Information. Reuters said it could not independently verify that financing report at the time.

In August 2026, Reuters reported that Nvidia was discussing an investment in a new round that could value Perplexity at more than $30 billion. That figure represented financing discussions rather than a confirmed completed valuation at the time of reporting.

The most important historical change is not the funding trajectory but the product trajectory:

Perplexity began by answering questions, expanded into deeper research, and is now attempting to automate parts of knowledge work itself.

Leadership

Perplexity remains closely associated with its founding team.

Aravind Srinivas: Co-Founder and CEO

Aravind Srinivas is Perplexity's co-founder and chief executive officer.

Before founding Perplexity, he worked in AI research, including roles associated with OpenAI and Google DeepMind. He remains the company's most prominent public spokesperson and is closely associated with its strategy around search, AI agents, and the future of information discovery.

Denis Yarats: Co-Founder and CTO

Denis Yarats is Perplexity's co-founder and chief technology officer.

Before Perplexity, he worked at Meta's Facebook AI Research organization and had experience with search technology. His current CTO role makes him one of the most important executives shaping Perplexity's technical direction.

Johnny Ho: Co-Founder and Chief Strategy Officer

Johnny Ho is a co-founder and chief strategy officer.

His role places him on the strategic side of Perplexity's expansion as the company moves into additional products, customer segments, and markets.

Andy Konwinski: Co-Founder

Andy Konwinski co-founded both Databricks and Perplexity.

Forbes reported in 2026 that he no longer holds an active operating position at either company and has shifted attention toward venture investment and research initiatives.

Ownership, Subsidiaries and Corporate Structure

Perplexity AI is a privately held company, not a publicly traded corporation.

It does not have a public stock symbol, and there is no publicly traded parent company above it. Ownership is distributed among founders, employees, and outside investors, but a complete current capitalization table is not publicly available.

Reported investors and backers have included Nvidia, SoftBank, Jeff Bezos, and other technology investors. Forbes has also reported backing from prominent AI researchers and technology figures.

Because Perplexity remains private, investors do not receive the same detailed ownership, subsidiary, segment, and governance disclosures that would be available in a public company's annual filings.

From an operating perspective, the company is easier to understand as a single AI platform with multiple product lines—Search, Research, Computer, Comet, Enterprise, and APIs—rather than as a conglomerate of independently reported subsidiaries.

Perplexity AI Strategy

Perplexity's current strategy can be summarized as a transition from search engine to AI work platform.

Several priorities define that strategy.

Move From Search to Action

The most important strategic shift is the move from answering questions to completing objectives.

Search retrieves and synthesizes information.

Research performs deeper investigation.

Computer attempts to perform work based on the information it gathers.

This Search → Research → Action progression is the clearest way to understand Perplexity's product direction.

Own More of the User Interface

Comet brings Perplexity directly into the browser.

Personal Computer extends it toward the operating system and local files.

Portable Computer moves parts of the agent workload onto the user's device.

Together, these products reduce Perplexity's dependence on users visiting a standalone search page.

Expand Enterprise AI

Enterprise products allow Perplexity to monetize professional and organizational research at higher value than ordinary consumer search.

The company is developing increasingly detailed enterprise roles, permissions, internal knowledge search, and administrative functionality.

Become AI Infrastructure for Developers

The API Platform extends Perplexity beyond its own applications.

If other companies build Perplexity's search or agent tools into their products, the company can participate in AI usage without controlling the final user interface.

Invest in Agent Infrastructure

Perplexity's 2026 product development increasingly emphasizes long-running and resource-intensive agents.

The company has published work on secure agent runtimes, agent memory, local-first execution, and agent security. It has also said it plans to use Nvidia's Vera CPUs for AI-agent workloads.

This suggests that Perplexity increasingly sees its future not only in producing better AI responses but in operating persistent systems that can perform work.

Key Challenges and Risks

Perplexity's expansion creates opportunities, but it also exposes the company to several significant risks.

Competition From Much Larger Technology Companies

Perplexity competes with companies that have substantially larger user bases, infrastructure budgets, distribution channels, and existing software ecosystems.

Google controls major search and browser platforms. OpenAI competes in AI assistants, search, agents, and browsers. Microsoft and Google have deep enterprise relationships.

Perplexity therefore needs to differentiate through product quality, speed, citations, agent capabilities, or user experience rather than relying on distribution scale.

Citation and Answer Reliability

Perplexity's citations make its responses easier to inspect, but they do not guarantee that every generated statement is accurate.

The system can still misunderstand a source, rely on weak evidence, or produce a conclusion that is not fully supported by the cited material.

This risk becomes more important as Perplexity moves from answering questions to taking actions.

Copyright and Data-Access Disputes

Perplexity faces ongoing disputes concerning how AI companies access and use content.

CNN filed a copyright lawsuit against Perplexity in May 2026, alleging unlawful use and distribution of its material. Perplexity disputed the allegations.

Reddit has also sued Perplexity over alleged data scraping. In July 2026, a federal judge allowed key parts of Reddit's case to continue after rejecting most of Perplexity's bid for dismissal.

These cases form part of a broader unresolved debate over AI search, copyright, scraping, licensing, and publisher compensation.

Privacy and Agent Permissions

The risk profile changes when AI gains access to files, applications, browser sessions, email, or company systems.

Perplexity's Enterprise products increasingly include role-based access control and administrative permissions, while local and desktop agents require explicit device-level access.

As Perplexity automates more actions, security failures or poorly designed permissions could have greater consequences than an incorrect search result.

High Infrastructure Requirements

Search, Research, and especially long-running agents require significant computing infrastructure.

Perplexity's plan to adopt Nvidia's Vera processors for agent workloads illustrates how important compute performance is becoming to the company's products.

This makes infrastructure availability, efficiency, and cost increasingly important to the company's ability to scale.

Regulatory Risk

AI-generated search is attracting regulatory attention.

In July 2026, German media regulators said that AI-generated summaries from products including Perplexity could fall under German media-law requirements, reflecting increasing scrutiny of how AI systems select, summarize, and distribute information.

Key Facts About Perplexity AI

1. Perplexity was founded in 2022, making it a relatively young company despite already competing across search, browsers, enterprise AI, and agent platforms.

2. Its core search experience combines web retrieval, generative AI, and citations, differentiating it from both traditional link-based search and completely uncited AI responses.

3. Perplexity uses multiple AI models rather than relying exclusively on one model provider.

4. Its product evolution follows a Search → Research → Action model, with Search answering questions, Research conducting deeper investigations, and Computer completing multi-step tasks.

5. Comet gives Perplexity its own AI-browser layer, allowing the company to compete for user activity before a person reaches a traditional search engine.

6. Personal Computer and Portable Computer extend Perplexity toward local and on-device AI agents, giving the platform access to environments beyond the web when users grant permission.

7. The company serves consumers, students, enterprises, and software developers, allowing the same core technology to support multiple revenue models.

8. Perplexity remains privately held, so detailed audited revenue, ownership percentages, and segment financials are not publicly disclosed in the same way they would be for a listed company.

9. The company was reportedly valued at about $20 billion in 2025, while investment discussions reported in August 2026 contemplated a valuation above $30 billion.

10. Perplexity's biggest strategic transition is from information retrieval to execution—a shift that also increases the importance of security, permissions, cost, and reliability.

Frequently Asked Questions

What is Perplexity AI?

Perplexity AI is an AI-powered search, research, and agent platform. Its core search product retrieves information from the web, generates conversational answers, and provides citations to the underlying sources.

Is Perplexity AI free?

Yes. Perplexity offers a free Standard plan. It also sells paid Pro, Max, Education Pro, Enterprise Pro, and Enterprise Max plans, while developer API usage is billed separately.

Is Perplexity a search engine or a chatbot?

It has characteristics of both, but AI-powered answer engine or AI search platform is more precise. It combines a conversational interface with web search, information retrieval, AI synthesis, and citations.

Does Perplexity use ChatGPT?

Perplexity is not simply a ChatGPT wrapper. It operates its own search and orchestration platform and can use AI models from multiple providers depending on the product, plan, and current model availability.

Who owns Perplexity AI?

Perplexity is privately held. Its ownership is distributed among founders, employees, and outside investors. A complete current capitalization table is not publicly available.

Who founded Perplexity AI?

Perplexity was founded in 2022 by Aravind Srinivas, Denis Yarats, Johnny Ho, and Andy Konwinski.

Who is the CEO of Perplexity?

Aravind Srinivas is Perplexity's co-founder and CEO.

How does Perplexity AI make money?

Perplexity makes money primarily through consumer subscriptions, enterprise subscriptions, developer API usage, and advanced AI-agent usage.

What is Perplexity Computer?

Perplexity Computer is an AI-agent product designed to perform multi-step tasks and workflows rather than only returning answers. It can combine research with tools, application connectors, coding, file creation, and other actions.

What is the difference between Perplexity Research and Computer?

Research focuses on investigating complex questions and producing detailed reports. Computer goes further by using tools and connected services to complete tasks and workflows.

Can Perplexity access files on a user's computer?

The normal search interface does not automatically have unrestricted access to local files. Perplexity's Personal Computer product can work with local files and applications when the user installs the required software and grants the relevant permissions.

Sources:

• Perplexity — Subscription Plan Guide

• Perplexity — Research Mode Documentation

• Perplexity — Computer Documentation

• Perplexity — Getting Started With Comet

• Perplexity — Personal Computer Documentation

• Perplexity Blog — Product and Research Updates

• Perplexity — Enterprise Pricing and Billing

• Forbes — Perplexity Company Profile