Palo Alto Networks agrees €4.2bn deal for AI startup Applied Compute
Palo Alto Networks has agreed an all-stock acquisition of Applied Compute, valued at $4.5 billion (€4.2 billion), to add tools for training and running open-weight AI models to its cybersecurity business. The company says the platform will help enterprises secure models in public clouds and sovereign environments; the deal remains subject to regulatory clearance and is expected to close in its fiscal third quarter of 2027.
Bottom line — Palo Alto forecasts more than $930 million (€1 billion) in annual recurring revenue from the deal by fiscal 2029.
Go deeper 7
-
Applied Compute specialises in post-training and inference: adapting models and running them in applications, according to Palo Alto Networks.
-
Applied Compute co-founder and CEO Yash Patil is to become Palo Alto’s co-chief AI officer after the deal closes, the company said.
-
Palo Alto CEO Nikesh Arora said the company wants to build and serve secure open-weight models as well as protect them.
-
Palo Alto plans to add its own compute capacity, with most of the spending going to GPUs and servers, CFO Dipak Golechha said, according to CSO Online.
-
Arora said open-weight models had grown from under 5% to more than 60% of token traffic over the past year, though he did not identify the source of those figures, CSO Online reported.
-
Greyhound Research analyst Sanchit Vir Gogia cautioned that a model supervisor built by the same team as the system it monitors may not provide an independent check.
-
Palo Alto reported first-quarter fiscal 2027 revenue of $3.44 billion (€3.2 billion), up 39% year on year, Channel Dive reported.