Intel and Google Deepen Collaboration to Advance AI Infrastructure with Xeon CPUs and Custom IPUs

Key Terms

infrastructure processing units (ipus) technical

Infrastructure Processing Units (IPUs) are specialized computer chips built to handle data-center tasks such as moving data, managing network traffic, securing communications and controlling storage, so central processors can focus on running applications. They matter to investors because IPUs can lower operating costs, boost performance and enable new cloud and AI services; think of them as traffic managers that keep a busy highway flowing so the main engines can run faster and more efficiently.

asic-based technical

ASIC-based means a product or system built around an application-specific integrated circuit — a custom-designed chip made to do one particular job very efficiently. For investors, ASIC-based designs matter because they can deliver much higher performance, lower energy use and lower unit costs than general-purpose components, but they also tie a company to specific technology choices and longer lead times, affecting competitiveness and supply risk.

heterogeneous ai systems technical

A heterogeneous AI system is a setup that combines different kinds of artificial intelligence models, software tools and hardware components so each part handles the tasks it does best — like an orchestra where different instruments play different roles. For investors it matters because this mix can improve performance, flexibility and resilience but also raise integration, cost and vendor-risk questions that affect a product’s speed to market, operating expenses and long‑term competitive edge.

ai inference technical

AI inference is the step where a trained artificial intelligence model uses its learned patterns to analyze new data and produce an output — for example, predicting a stock trend, flagging a medical image, or generating text, much like using a recipe to cook a meal. It matters to investors because inference determines real-world performance, speed, and cost of AI features, affects user experience and scalability, and influences operating expenses, regulatory compliance, and competitive advantage.

latency-sensitive inference technical

Latency-sensitive inference is when a computer model must produce predictions or decisions almost instantly, because delays would harm the outcome. Think of it like a cashier scanning items or a traffic light changing on time: if the response is slow, customers get frustrated and processes fail. For investors, this matters because systems that need near-instant results affect user experience, revenue opportunity, infrastructure cost and competitive advantage in real-time markets.

hyperscale ai environments technical

Large, purpose-built computing setups designed to train and run advanced artificial intelligence models at very high speed and volume, combining vast racks of specialized chips, fast storage and networking, and software that coordinates them like a factory assembly line. For investors, they matter because they require heavy upfront spending but can deliver faster product development, lower unit costs and recurring revenue for companies that control them, creating potential scale advantages or capital risks.

data center architectures technical

Data center architectures are the overall layouts and designs that determine how servers, storage, networking, cooling, and power systems are organized and managed to run large-scale computing operations. Like the floor plan and workflow of a factory or library, the architecture affects speed, cost, energy use, and reliability — key factors investors watch because they influence a company’s operating expenses, capacity to grow, and risk of downtime.

NEWS HIGHLIGHTS:

• Intel® Xeon® processors to continue powering Google Cloud infrastructure across AI, inference and general-purpose workloads
• Expanded co-development of custom ASIC-based infrastructure processing units (IPUs) to improve efficiency, utilization and performance at scale
• Collaboration reinforces the central role of CPUs and IPUs in modern, heterogeneous AI systems

SANTA CLARA, Calif.--(BUSINESS WIRE)-- Intel Corporation (NASDAQ: INTC) and Google today announced a multiyear collaboration to advance the next generation of AI and cloud infrastructure, reinforcing the critical role of CPUs and custom infrastructure processing units (IPUs) in scaling modern, heterogeneous AI systems.

As AI adoption accelerates, infrastructure is becoming more complex and heterogeneous, driving increased reliance on CPUs for orchestration, data processing and system-level performance. Through this collaboration, Intel and Google will align across multiple generations of Intel® Xeon® processors to improve performance, energy efficiency and total cost of ownership across Google’s global infrastructure.

AI doesn’t run on accelerators alone - it runs on systems. And CPUs are at the core of those systems.

Google Cloud continues to deploy Intel Xeon processors across its workload-optimized instances, including the latest Intel Xeon 6 processors powering C4 and N4 instances. These platforms support a broad range of workloads—from large-scale AI training coordination to latency-sensitive inference and general-purpose computing.

In parallel, Intel and Google are expanding their co-development of custom ASIC-based IPUs. These programmable accelerators offload networking, storage and security functions from host CPUs - improving utilization, increasing efficiency and enabling more predictable performance across hyperscale AI environments.

IPUs are a critical component of modern data center architectures. By handling infrastructure tasks traditionally managed by CPUs, they unlock greater effective compute capacity and allow cloud providers to scale more efficiently without increasing overall system complexity. Together, Xeon CPUs and IPUs form a tightly integrated platform balancing general-purpose compute with purpose-built infrastructure acceleration to deliver more efficient, flexible and scalable AI systems.

Driving Performance and Efficiency at Scale

“AI is reshaping how infrastructure is built and scaled,” said Lip-Bu Tan, CEO of Intel. “Scaling AI requires more than accelerators - it requires balanced systems. CPUs and IPUs are central to delivering the performance, efficiency and flexibility modern AI workloads demand.”

“CPUs and infrastructure acceleration remain a cornerstone of AI systems—from training orchestration to inference and deployment,” said Amin Vahdat, SVP & Chief Technologist, AI Infrastructure, Google. “Intel has been a trusted partner for nearly two decades, and their Xeon roadmap gives us confidence that we can continue to meet the growing performance and efficiency demands of our workloads.”

Building the Foundation for the Next Wave of AI

The expanded collaboration reflects a shared commitment to advancing open, scalable infrastructure for the AI era. By combining general-purpose compute with purpose-built infrastructure acceleration, Intel and Google are enabling a more balanced approach to AI system design - one that improves utilization, reduces complexity and scales more efficiently.

Together, the companies are strengthening the foundation for the next generation of AI-driven cloud services—supporting continued innovation across enterprises, developers and users worldwide.

About Intel

Intel (Nasdaq: INTC) designs and manufactures advanced semiconductors that connect and power the modern world. Every day, our engineers create new technologies that enhance and shape the future of computing to enable new possibilities for every customer we serve. Learn more at intel.com.

View source version on businesswire.com: https://www.businesswire.com/news/home/20260409055389/en/

Intel Media Relations
ContactPR@Intel.com

Source: Intel Corporation