Perplexity Launches Autonomous AI 'Computer'
AI startup Perplexity has launched "Computer," a multi-model AI system designed to autonomously create and execute entire digital workflows. The platform can orchestrate complex business tasks like planning, research, and execution, marking another step toward fully autonomous enterprise operations.
Founded in 2022 by former AI researchers from OpenAI, Google, and DeepMind, Perplexity AI has rapidly grown, reaching a valuation of over $1 billion by April 2024 and processing 780 million queries in May 2025. The company is led by co-founder and CEO Aravind Srinivas, a computer science PhD from UC Berkeley. Perplexity's "Computer" is designed to act as a "general-purpose digital worker" by using multiple AI models to execute complex, multi-step tasks from start to finish. It can handle workflows that run for hours or even months, a significant leap from the conversational question-and-answer capabilities of standard chatbots. Instead of relying on a single AI, "Computer" orchestrates 19 different specialized models. For instance, it might use Claude Opus for high-level reasoning, Gemini for in-depth research, and Nano Banana for image generation, assigning the best model to each sub-task in a workflow. This new platform can interact with software interfaces like a human would, allowing it to perform tasks like building an application, integrating with tools such as Gmail and Slack, or conducting data analysis. Each task is executed within a secure cloud sandbox that has browser and filesystem access. The system is a direct competitor to emerging autonomous agent platforms like OpenClaw. It is initially available to subscribers of Perplexity's "Max" tier, which costs $167 per month. The company has secured significant funding from major investors including Jeff Bezos, Nvidia, and Databricks. By September 2025, Perplexity's valuation had reportedly reached $20 billion.
