Nvidia is in talks to invest in Perplexity AI at a valuation exceeding $30 billion, a more than 50% jump from the company's previous $20 billion round, according to The Information. The multi-billion-dollar round caps a four-year run that would make Perplexity one of the highest-valued independent players in AI search - and it signals how deeply chip sales now depend on AI application growth.
The deal wasn't originally structured as an equity investment. Nvidia initially planned to spend billions acquiring technology licenses from Perplexity and absorbing some of its employees, the report said. Those talks pivoted to a conventional equity path. Nvidia CEO Jensen Huang has said he uses Perplexity daily, but the logic behind the investment runs deeper.
The product that changed the numbers
Perplexity built its name on conversational search - asking a question and getting a synthesized answer with sources rather than a list of links. That model earned it the "Google killer" label, and Google responded by accelerating its own AI search efforts.
But the valuation jump is driven by Perplexity Computer, a cloud-based AI agent product for professionals launched February 25, 2026 that autonomously executes complex tasks on computers. It competes directly with agent offerings from OpenAI and Anthropic.
The product has reshaped the company's trajectory. Perplexity's annualized revenue rose from under $250 million at the start of 2026 to more than $750 million by August. Monthly active users across its search and agent tools have surpassed 100 million, including tens of thousands of enterprise customers.
Nvidia's compute logic
The investment thesis is straightforward: more tasks executed by Perplexity Computer means more compute consumed - and more chips sold. Perplexity has already agreed to use Nvidia's Vera CPUs for agentic AI queries and signed a chip usage agreement with CoreWeave, a cloud provider heavily backed by Nvidia. The two companies now meet multiple times a week on hardware and software development.
This "invest in application entry points" playbook has precedent. Nvidia committed up to $100 billion in OpenAI in 2025, and joined Microsoft in a funding round for Anthropic with a commitment of up to $10 billion. PitchBook counted 67 venture deals with Nvidia in 2025, many of which involved portfolio companies spending their proceeds on Nvidia GPUs.
"This investment is both a financial bet and an ecosystem lock-in," and Perplexity sits at one of the Nvidia-aligned Nemotron Alliance's core nodes, built with Nvidia in March 2026, which includes Mistral, Cursor, and Thinking Machines to share compute and data.
Capital growth and valuation risk
Perplexity's trajectory is measured in quick steps: a $3.1 million seed round in September 2022 before product launch, a $25.6 million Series A at a $150 million valuation in March 2023, and a series of raises through 2024 and 2025 that pushed its valuation to $20 billion. Cristiano Ronaldo also joined as an investor and global brand ambassador in 2025. By early 2026, another round took it to approximately $21.2 billion.
The $30 billion target price implies a price-to-sales ratio of roughly 40x against $750 million in annualized revenue. For comparison: OpenAI's $500 billion valuation against about $20 billion revenue works out to roughly 34x, and Anthropic is valued at $85 billion for around $5 billion in revenue - about a 20.5x multiple.
A more structural concern is that Perplexity's Sonar model fine-tunes Meta's Llama and relies on multi-model routing, pulling its strongest answers from third-party models by OpenAI, Anthropic, Google, and xAI. If suppliers raise prices or restrict access, Perplexity's cost structure comes under pressure as a customer and competitor at the same time. If it can lean on Nvidia's Nemotron ecosystem instead, its $30 billion valuation has a more stable basis.
Why this matters for finance professionals
Finance teams budgeting for AI tools now have a clearer measure of what AI search and agent productivity tools are worth. Investors watching the earnings of not just Perplexity but the entire AI compute supply chain have a direct tell: agentic AI products have tripled revenue in eight months - sustainable compound growth goes into the 40x P/S that justifies the price. That depends on perpendicular model costs and chips. The Nvidia investment keeps an inventory of committed customers active, which makes their spending more predictable to forecast.
For finance professionals considering applications, this isn't just a head-of-search product but a endpoint that ties into the existing compute dependency - likely through Perplexity AI training courses built for practical workflows. If agentic costs scale efficiently with usage, it could part of the next year's budgeting. That remains a test - not a guaranteed outcome, but a new material risk in the AI narrative.
Your membership also unlocks:
