Skip to content
BoultonSubscribe
MenuClose

Palo Alto Networks wants to run the AI it secures

The $4.5 billion Applied Compute deal tests whether a cybersecurity vendor can win over the teams deploying AI models.

Palo Alto Networks CEO Nikesh Arora
Photo: Godofredo A. Vásquez / AFP via Getty Images

Palo Alto Networks wants to be more than the company protecting enterprise AI. It wants a role in running it, too.

On November 18, the cybersecurity company agreed to acquire Applied Compute for $4.5 billion in stock. Applied Compute specializes in adapting and deploying open-weight AI models, giving Palo Alto a potential foothold beyond cybersecurity and into AI infrastructure. The deal is expected to close in Palo Alto's fiscal third quarter of 2027.

The catch

Selling security software gives Palo Alto a relationship with the security team. Winning model deployments means convincing AI and infrastructure teams to build around its tools. For customers already deploying models elsewhere, built-in security may be valuable without being enough to justify switching platforms.

The competitive picture

Microsoft already offers managed deployment for open models through Microsoft Foundry. Palo Alto is betting it can win customers that want to customize and run models in public-cloud or sovereign environments, with security built into the process.

If it works, Palo Alto could capture a larger share of what customers spend deploying and running AI models.

Get Boulton in your inbox

Every Thursday

By subscribing, you accept our Privacy Policy.