AI is fundamentally altering the requirements for network infrastructure, yet many organizations, including neocloud providers, are not adequately prepared. A recent study by analyst firm Omdia highlights that while many GPU-as-a-service providers have expanded their compute capabilities to meet AI workload demands, their networking infrastructure is increasingly becoming a limiting factor.
Neocloud Providers Under Scrutiny
Neocloud operators, which emerged to capitalize on the growing need for GPU acceleration in AI, serve a range of clients, including major hyperscalers like Microsoft. The performance of AI applications is now heavily reliant on these providers’ ability to efficiently process and transfer data across diverse environments. However, Omdia notes that the networking capabilities of these neoclouds vary significantly, with some having origins in cryptocurrency mining and others in content distribution or web hosting.
Urgent Need for Networking Upgrades
As the dependency on robust networking grows, many neocloud operators are actively seeking partnerships, acquisitions, or infrastructure development to enhance their capabilities. Camille Mendler, Telco B2B Research Director at Omdia, emphasizes that “network infrastructure will make or break neoclouds.” She asserts that low latency, resilient, and secure connectivity is essential for success, especially given the complexities of AI workload sovereignty.
Lumen’s Call to Action
Global network provider Lumen is also recognizing this shift. CEO Kate Johnson has reached out to enterprise leaders, questioning whether their networks are prepared for AI demands and advocating for necessary upgrades. Johnson likens traditional networking to plumbing, stating that in an AI-driven enterprise, the network functions more like a nervous system, coordinating and controlling data flow. She highlights that AI systems require constant data movement across clouds, data centers, and edge endpoints, necessitating adaptable and scalable networks.
The Rise of Autonomous Data Traffic
Johnson points out that over 50 percent of internet traffic is now generated by autonomous agents and bots, according to Imperva’s 2025 Bad Bot Report. This shift underscores the urgent need for networks to be fully adaptable, programmable, and consumption-based, similar to cloud infrastructure. She urges enterprise leaders to ensure their networks align with the future they envision.
This article was produced by NeonPulse.today using human and AI-assisted editorial processes, based on publicly available information. Content may be edited for clarity and style.
