Charles Lyons-Burt 
As federal agencies accelerate the adoption of artificial intelligence to support mission-critical operations, industry is facing a new challenge: delivering the infrastructure behind AI at the speed government demands.
From AI factories and next-generation data centers to GPU shortages and procurement reform, success increasingly depends on collaboration across government and technology providers.
In a conversation with ExecutiveBiz, Maryam Emdadi, vice president of hardware sales at Carahsoft, discusses why AI factories have become a top priority, the supply chain challenges shaping the market and how stronger partnerships are helping agencies deploy AI faster.
We last spoke with Emdadi in Aug. 2025 about lifecycle planning, cloud repatriation and AI-ready PCs.
ExecutiveBiz: What are the biggest trends you’re seeing this year?
Maryam Emdadi: The theme of this year is helping the government move fast. This administration has embraced a willingness to move quickly and disrupt traditional processes, which is an important mindset for adopting emerging technologies. Our responsibility as industry partners is to keep pace and help agencies achieve their goals.
AI factories are also a big theme this year. We’ve spent significant time strategizing how to deliver AI factories by expanding our portfolio and ensuring we have the technologies, partners and alternatives needed to support these large-scale deployments.
ExecutiveBiz: What are the components of an AI factory?
Emdadi: An AI factory is the infrastructure that powers an AI data center, and we’re seeing these environments being built across the country. For example, NVIDIA recently partnered with the Naval Postgraduate School to launch an AI factory, while initiatives such as the Golden Dome project demonstrate how AI infrastructure is becoming critical to national defense.
An AI factory is much more than GPUs. It includes servers, storage, networking, power and cooling, cloud services, cybersecurity and operational management. Every component has to work together to support AI workloads at scale.
That has fundamentally changed how solutions are sold. Organizations aren’t buying individual servers anymore. They’re investing in AI outcomes. Every deployment requires multiple technology partners working together to deliver a complete solution around a specific mission or AI workflow.
ExecutiveBiz: What’s making these deployments difficult today?
Emdadi: The biggest challenge is supply. Demand for GPUs, memory and other critical hardware continues to outpace availability, while efforts to expand U.S.-based manufacturing are still ramping up. As a result, prices are increasing 30 to 80 percent every single month and the problem isn’t expected to ease until 2029.
Those challenges are especially difficult in the federal market because procurement timelines are much longer than they are in the commercial sector. A commercial organization can make a purchasing decision immediately, while a government acquisition may take months. By the time procurement is complete, pricing may have risen significantly and product availability may have shifted.
For national security programs, that’s a real concern. The government needs access to AI infrastructure quickly, and procurement processes need to evolve so agencies can secure technology before supply becomes even more constrained.
ExecutiveBiz: How are agencies and industry adapting?
Emdadi: A major focus for us has been helping both our partners and government customers navigate these market conditions. That means educating agencies about ongoing price increases and supply constraints while encouraging earlier purchasing decisions whenever possible.
We’ve also focused on diversifying our AI factory ecosystem so agencies have greater flexibility when shortages affect specific vendors or technologies. At the same time, we’re working with organizations such as GSA and OMB as they modernize procurement approaches so agencies can acquire AI infrastructure more efficiently.
Ultimately, the goal is simple: help government agencies move faster without sacrificing mission requirements.
An AI factory is much more than GPUs. It includes servers, storage, networking, power and cooling, cloud services, cybersecurity and operational management. Every component has to work together to support AI workloads at scale.
ExecutiveBiz: How are partnerships changing the way AI solutions are delivered?
Emdadi: AI has made collaboration essential. Every major opportunity now involves multiple OEMs working together because no single vendor provides the entire solution.
We’ve long believed in a ‘better together’ approach, and today that’s become the industry standard. Every engagement combines infrastructure, networking, cloud, software and services from multiple technology providers.
That’s where the channel becomes especially valuable. Rather than asking agencies to manage contracts with dozens of vendors, we help bring those technologies together into a single, integrated solution. We coordinate the partners, simplify procurement, support implementation and help agencies move from planning to deployment much faster.
As AI initiatives continue to grow in scale and complexity, success will depend not only on innovative technology, but on strong partnerships that make acquisition, integration and deployment as seamless as possible.
