Extreme Networks wants the network to run AI, not just be run by it

What changed

Extreme Networks unveiled new AI capabilities that shift the role of the network from simply being managed by AI to actively running AI tasks on the network itself. Their AI Summit in Amsterdam showcased Agent ONE Coworker, an early model that uses spare compute resources from access points and switches. These devices form “Exchange zones” which act as localized AI processing hubs. This means on-premises network hardware is no longer just data carriers or management end points but integral parts of AI infrastructure.

Why builders should care

Network operators and IT teams operate under tighter performance and security demands while wanting faster, smarter AI capabilities at the edge. By leveraging unused compute power in existing network gear, Extreme Networks lowers reliance on cloud AI or separate AI hardware. This reduces latency and bandwidth costs for AI inference and decision-making processes tied directly to network operations. Builders gain a practical way to integrate AI closer to data sources with less infrastructure overhead.

The practical takeaway

For teams managing distributed networks, this approach means better real-time AI-driven insights and automation without installing new AI servers. AI workloads related to network health, security, and user experience can run locally, speeding up detection and response times. It also opens chances to optimize network assets, extracting value from otherwise idle compute cycles. However, operators will need to rethink capacity planning and ensure AI tasks do not interfere with core networking functions.

What to watch next

Tracking how Agent ONE Coworker evolves and how widely Exchange zones deploy will show if this edge AI vision gains traction. IT pros should watch how well this local AI integrates with cloud AI and existing management tools. Another critical question is how security and privacy are maintained when AI workloads multiply within network hardware. Finally, the cost-benefit balance of turning network devices into AI nodes will determine if this model spreads beyond niche early adopters.

AI Quick Briefs Editorial Desk