Rethinking AI connectivity across data center, cloud and metro environments

Almost every enterprise has an AI roadmap. Models are being trained, workloads are scaling and real-time analytics are informing decisions that once took days.

But between the ambition and the outcome, something often breaks. And it’s almost always the network.

AI is fundamentally changing how data moves across distributed environments. As workloads span data centers, clouds and metro locations, the connectivity between them has become a critical dependency and a common bottleneck to scaling AI.

To keep pace, enterprises need connectivity that reaches key data centers, clouds and metro environments, supports private high-capacity traffic flows and scales on demand.

That’s why Lumen is expanding its data center and metro footprint across 70+ third-party, cloud on-ramp-ready facilities and 16 major U.S. markets*—bringing high-performance, AI-ready connectivity closer to where data is created, processed and exchanged.

Legacy networks are holding AI back

Enterprise networks were built for Cloud 1.0—an era defined by predictable, north-south traffic, where users accessed data and applications from centralized environments.

That model no longer applies.

AI, machine learning and real-time analytics are driving massive volumes of east-west data movement. Training workloads move large datasets between distributed GPU clusters and storage across multiple locations. Applications require low-latency connectivity between environments. Hybrid and multicloud strategies demand consistent, private access to multiple providers without the variability of the public internet.

However, most enterprise networks remain static, manually provisioned and fragmented. They are designed for simpler, centralized traffic patterns—not the distributed, real-time data movement AI requires. The results:

• Data bottlenecks that slow AI model training and delay time to value—1 PB of data can take approximately 222 hours to transfer at 10G versus approximately six hours at 400G¹
• Inconsistent performance impacting AI workloads and application reliability—82% of enterprises report performance issues with AI workloads²
• Limited cost control, where static provisioning leads to overbuild or underutilization instead of aligning spend to demand

What AI demands from modern connectivity

The shift businesses must take doesn’t just mean adding more bandwidth. It centers on rethinking connectivity for an AI-driven Cloud 2.0 world—where networks must support continuous, high-volume data movement across distributed environments.

To support this shift, enterprises need four core capabilities:

Dynamic scale aligned to demand
AI workloads are unpredictable. Connectivity must scale up or down on demand—from 1G to 400G and beyond—without overprovisioning. This requires a consumption-based model that aligns capacity, cost and usage in real time, enabling cost predictability and control.

Expansive reach across environments
Modern workloads span colocation facilities, cloud regions, metro sites and on-premises environments. Connectivity must seamlessly unify these endpoints through a single network fabric, eliminating fragmented providers and simplifying how enterprises connect and scale.

Predictable, private performance
AI and real-time applications depend on consistent, low-latency performance. Private, SLA-backed connectivity is becoming essential to support reliability and security, while removing the variability and congestion risks of public internet routing.

Programmable self-service control
AI operates at a faster pace than traditional networking. Teams need API-driven, programmable, self-service control to provision, scale and monitor connectivity in minutes—not weeks—without manual processes.

For Lumen, these capabilities define what Cloud 2.0 connectivity looks like in practice: scalable consumption, deep network reach, private performance and digital control via a unified platform for AI.

How Lumen delivers AI-ready connectivity

To achieve this, Lumen is expanding its data center and metro footprint—bringing high-capacity, on-demand connectivity closer to where AI workloads live, move and scale.

Through this expansion, Lumen delivers private connectivity across 70+ third-party, cloud on-ramp-ready data centers and 16 major U.S. metro markets*—with speeds up to 400 Gbps at key cloud data centers and up to 100 Gbps across regional data centers, campuses and edge locations.

This expanded footprint reduces the distance between compute, storage and users—helping minimize latency, accelerate data movement and enable more efficient placement of AI workloads across distributed environments. Enterprises can connect across data centers, clouds and metro locations through a single, private network, rather than stitching together fragmented providers.

Built on the Lumen-owned, deeply peered fiber infrastructure, the network delivers consistent, low-latency performance with SLA-backed reliability, enabling AI, cloud and data-intensive workloads to move predictably and at speed—without exposure to the public internet.

Bringing connectivity under your control

What differentiates Lumen is not just the network, but how connectivity is delivered and consumed. Through an on-demand, Network-as-a-Service (NaaS) model powered by the fully digital, API-driven Lumen Connect℠ platform, organizations can provision bandwidth in minutes, scale capacity up or down in real time and manage connectivity across environments through a single, self-service interface.

This model enables cost predictability, consumption governance and operational simplicity, while giving enterprises the agility required to support fast-moving AI initiatives.

Together, the Lumen expansion across data centers and metro environments, its high-performance network and its digital platform create a unified, Cloud 2.0-ready foundation—purpose-built to support the scale, performance and flexibility required for AI-driven workloads.

The result is a more agile connectivity model that enables enterprises to move data faster, adapt to changing demand and turn AI ambition into real-world outcomes.

AI readiness depends on how data moves

As modern workloads become more distributed and data flows more dynamic, connectivity is no longer a supporting function. It’s the trusted foundation for performance, insight and competitive advantage.

The organizations that succeed will be those that reduce friction between data, clouds and metro environments—enabling faster scaling, greater agility and more consistent outcomes.

Lumen is focused on investing in network footprint, performance and delivery. The goal is to bring high-capacity, low-latency connectivity closer to where data is created, processed and consumed, and to support the demands of AI at scale.

See how Lumen helps enterprises move AI workloads across data center, cloud and metro environments with private connectivity built for Cloud 2.0.

_{¹Lumen Technologies, Cloud 2.0: because AI won’t run on yesterday’s internet, 2025.}

_{²Flexential, 2024 state of AI infrastructure, September 2024.}

_{*This represents phase one of the broader Lumen expansion, with additional data centers and metro markets expected to be enabled with these capabilities in future phases.}

_{This content is provided for informational purposes only and may require additional research and substantiation by the end user. In addition, the information is provided "as is" without any warranty or condition of any kind, either express or implied. Use of this information is at the end user's own risk. Lumen does not warrant that the information will meet the end user's requirements or that the implementation or usage of this information will result in the desired outcome of the end user. All third-party company and product or service names referenced in this article are for identification purposes only and do not imply endorsement or affiliation with Lumen. This document represents Lumen products and offerings as of the date of issue. Services not available everywhere. Lumen may change or cancel products and services or substitute similar products and services at its sole discretion without notice. © 2026 Lumen Technologies. All Rights Reserved.}