Skip to header Skip to main navigation Skip to main content Skip to footer
Scott Lawson
Your next upgrade lives here

Palo Alto Networks Bets 4.5 Billion Dollars on Applied Compute to Secure Open-Weight AI

Palo Alto Networks said on Wednesday that it has agreed to acquire Applied Compute, a young artificial intelligence startup built by former OpenAI researchers, in an all-stock deal valued at 4.5 billion dollars. The cybersecurity company announced the definitive agreement on November 18, alongside fiscal first-quarter results that came in well ahead of Wall Street expectations, and said it will give further details on its earnings call scheduled for 1:30 p.m. Pacific Time the same day.

Applied Compute, founded in 2025 by Yash Patil, Rhythm Garg and Linden Li, offers a platform for the post-training and inference of open-weight AI models. Its tools are designed to make it easier for companies to train, serve and improve models tuned to their own data, and include an autoscaling inference engine and a model router called Shift-1. The startup raised 350 million dollars earlier this month at a 3.25 billion dollar valuation in a round led by Elad Gil, with Nvidia among the participants, and said at the time that its annualized revenue had passed 100 million dollars.

Palo Alto Networks is pitching the deal as a bet that the future of enterprise AI is hybrid. The company argues that a large share of enterprises will run open-source and open-weight models tuned to their own data, cost and control needs, alongside closed frontier models, and that no large provider is yet delivering those open models securely. With Applied Compute, customers would be able to train, adapt and deploy open-weight models with security built in, either in the public cloud or in sovereign environments intended to help with compliance.

"Every enterprise is going to run a mix of open and closed models, and every one of those models needs to be secured," Chairman and Chief Executive Nikesh Arora said. "Applied Compute has built the best post-training and inference platform we've seen, and its research team is exceptional." Arora added that Palo Alto Networks had seen the platform firsthand as a prospective customer, and that its Unit 42 Frontier AI Assessment had confirmed the startup's strong security foundations. Patil said joining Palo Alto Networks would put security at the center of Applied Compute's products and accelerate its traction with the world's largest organizations.

Under the agreement, the Applied Compute team will join Palo Alto Networks, and Patil will become the company's Co-Chief AI Officer once the transaction closes. The research team is also expected to help Palo Alto Networks embed AI across its own offerings, training its proprietary large and small models. The deal is subject to customary closing conditions, including regulatory clearance, and is expected to close in Palo Alto Networks' fiscal third quarter of 2027, roughly a year and a half after the startup's latest funding round valued it at 3.25 billion dollars.

The acquisition landed on a strong day for the buyer. For the quarter ended October 31, Palo Alto Networks reported adjusted earnings of 1.05 dollars per share, up from 93 cents a year earlier, on revenue of 3.44 billion dollars, up 39 percent year over year and ahead of the 3.31 billion dollars analysts expected. The company said it is materially raising its full-year guidance, and its shares rose in late trading. Even so, the Applied Compute deal is the longer story: a security vendor paying 4.5 billion dollars in stock for a two-year-old model platform is a clear signal that securing AI is no longer enough. The vendors want to help run it, too.

Artificial Intelligence
Cybersecurity
Big Tech
Startups
AI Infrastructure
Technology

Copyright © 2026 Rocky Mountain Madman LLC - All rights reserved

Developed and Designed by Rocky Mountain Madman LLC