Palo Alto Networks to Acquire Applied Compute for $4.5 Billion to Expand AI Security

Summary
Palo Alto Networks plans to buy Applied Compute for $4.5 billion to expand enterprise AI security. CEO Nikesh Arora says the company aims to develop models that supervise AI systems and address threats such as prompt injection.
Key points
- Palo Alto Networks plans to acquire Applied Compute for $4.5 billion.
- Palo Alto Networks selected Applied Compute from nine providers for its expertise in post-training open-weight models.
- The company plans to apply Applied Compute's capabilities to more than 35 specialized use cases targeting AI-enabled threats.
- Palo Alto Networks says it intends to offer enterprises secure AI training, inference and supervision capabilities.
- CEO Nikesh Arora said static controls may not anticipate autonomous agents' behavior, including attempts to circumvent controls.
- Palo Alto Networks expects the acquisition to contribute more than $1 billion in next-generation security annual recurring revenue by 2029.
Article Details
- Event Type
- Corporate acquisition
- Impact
- Palo Alto Networks will acquire Applied Compute for $4.5 billion to expand its ability to train, deploy, and secure enterprise AI models. The company expects the acquisition to add more than $1 billion in next-generation security annual recurring revenue by 2029; this is a projection, not a reported realized impact.
People
Vendors
Applied ComputeAI security must evolve toward models trained to understand and supervise other AI systems. Buying Applied Compute gives Palo Alto Networks the technical expertise to pursue that strategy while creating aPalo Alto NetworksPalo Alto Networks will purchase for $4.5 billion a startup led by an ex-OpenAI technical staff member to train, deploy and secure enterprise AI models.
Products
Cortextoken expenditure. Arora expects growing AI traffic to increase demand for Prisma AI, network security, Cortex, observability, identity protection and Applied Compute's training and inference services.Prisma AI2% to 5% of enterprise token expenditure. Arora expects growing AI traffic to increase demand for Prisma AI, network security, Cortex, observability, identity protection and Applied Compute's training and
