Perplexity Launches 'Digital Employee' AI
Perplexity just launched Perplexity Computer, a multi-model AI system designed to act as a 'digital employee.' Unlike single-model copilots, it orchestrates entire business workflows end-to-end, from data synthesis to approvals and communication, signaling a major shift from task execution to full workflow automation.
Perplexity's valuation has skyrocketed, jumping from $520 million in January 2024 to $9 billion by December 2024. The company, founded in August 2022, has attracted funding from prominent investors including Nvidia, SoftBank Vision Fund 2, and Jeff Bezos. The system is driven by a strategy of "multi-model orchestration," a significant departure from single-model copilots. Perplexity Computer uses an internal framework called OpenClaw to decompose a high-level goal into subtasks, then routes each piece of work to the best-suited model from a pool of 19 specialized AIs, including models from Google, OpenAI, and Anthropic. CEO Aravind Srinivas previously held research positions at OpenAI, Google Brain, and DeepMind after earning a Ph.D. from UC Berkeley and degrees from IIT Madras. He co-founded Perplexity with Denis Yarats (formerly of Meta AI), Johnny Ho, and Andy Konwinski. Unlike task-oriented assistants, Perplexity Computer is designed to execute long-running, autonomous projects that can last for hours, weeks, or even months. Use cases involve delegating entire workflows, such as market research report generation, competitive intelligence dashboard creation, or even coding and deploying applications from a single natural-language prompt. The offering signals a distinct move toward enterprise and prosumer users, with access available through paid subscriptions like the Pro plan at $20/month and a Max tier at $200/month. This pricing is based on usage credits, a shift away from a flat-rate or ad-supported model. The launch comes as the market for agentic AI is projected to expand dramatically, with forecasts suggesting growth from approximately $7 billion in 2025 to over $93 billion by 2032. This shift reflects a broader enterprise trend away from simple task automation and toward delegating complex, multi-step decision-making processes to autonomous systems.
