EMA Research Examines Enterprise Strategies for Supporting AI Inference at the Network Edge
A survey of 319 IT professionals found that only 43% of businesses expect their plans for running AI at the network edge to be completely successful.
LAFAYETTE, Colo., Nov. 17, 2026 /PRNewswire/ -- Enterprise Management Associates (EMA™), a leading IT research and consulting company, has published a new research report called Enterprise Strategies for AI Inference at the Network Edge. The report was written by Shamus McGillicuddy, EMA's vice president of research for network infrastructure and operations.
As businesses use more artificial intelligence (AI), many are looking at edge inference. This means running AI programs closer to where data is created. Businesses can do this by placing trained AI models in nearby data centers, cloud regions, or local edge systems. This can reduce delays, improve data security and privacy, and make it possible to process information in real time.
Edge inference allows businesses to use AI for tasks that central cloud systems may not handle well enough. Examples include making decisions in real time at remote factories and resource extraction sites, improving customer experiences in shops with AI, and using AI in physical machines such as self-driving vehicles and robots.
But supporting these demanding tasks takes more than installing AI models. Business networks, computer systems, and IT teams must be ready to handle applications that are especially affected by network congestion, delays, and other performance problems.
"Edge inference workloads will push enterprises to make significant upgrades and enhancements to network infrastructure at the data plane, control plane, and management plane," McGillicuddy said. "Network teams also need to manage security risk and consider compliance and data sovereignty requirements."
The new EMA report is based on a survey of 319 IT professionals. It looks at how businesses are planning to use edge inference and the problems they face with their technology and daily operations. The main findings include:
Only 43% of those surveyed expect their edge inference plans to be completely successful. IT executives were generally more positive than the technical teams responsible for putting the plans into action. This suggests a gap between what leaders expect and what teams can achieve in practice.
Only 18% run all their edge inference workloads at the remote locations where the data is created. Instead, 64% use a mix of edge locations and cloud systems, including public cloud services and data centers. Security rules, costs, and data sovereignty requirements are the main factors that affect where workloads run.
Only 40% of IT professionals are completely confident that their network systems are ready to support edge inference workloads. Network teams are making several improvements to prepare. For example, 97% are updating their SD-WAN systems, 97% are updating their DDI (DNS, DHCP and IPAM) platforms, and 96% are updating their certificate and identity management technology.
The three most common business goals for edge inference projects are making operations more efficient, improving safety and risk management, and making better decisions using information available in real time.
The report looks at the choices businesses are making about their technology, networks, and IT operations as they start using edge inference. It also explains early signs of success and what IT teams should consider when preparing to support these increasingly demanding AI tasks.
BlueCat and Scale Computing sponsored this independent research.
A full analysis of the findings is available in the report, Enterprise Strategies for AI Inference at the Network Edge.
EMA will hold a free webinar on December 2. Shamus McGillicuddy will share the main findings and recommendations from the research.
About EMA
Founded in 1996, EMA is a leading IT research and consulting company. It provides useful information and advice about the changing world of technology. Through independent research, market analysis, and reviews of technology providers, EMA helps businesses make well-informed technology decisions. Its team of analysts combines practical experience with a strong understanding of industry best practices and new technology solutions from vendors. This helps clients reach their business goals. Learn more about EMA's research, analysis, and consulting services at www.enterprisemanagement.com and follow the company on X and LinkedIn.
Media Contact:
Raleigh Gould
Enterprise Management Associates
[email protected]
View the original content and download multimedia: https://www.prnewswire.com/news-releases/ema-research-examines-enterprise-strategies-for-supporting-ai-inference-at-the-network-edge-302938616.html



