Extreme Connect 26: Agent ONE Takes Forward Network AI
Why It Matters
Embedding AI reasoning and execution into the network itself promises faster issue resolution, reduced manual effort, and a strategic shift toward autonomous enterprise infrastructure, raising the competitive bar for networking vendors.
Key Takeaways
- •Agent ONE Coworker launches July 2026, delivering proactive, ambient network assistance.
- •Four‑layer AI stack adds reasoning, context, SOPs, and execution within governance.
- •Agent ONE Operator will run continuously, automating tasks without human prompts.
- •Extreme Exchange marketplace enables custom industry‑specific AI skills for networks.
- •Analysts see Extreme’s move as leap beyond AI copilots to autonomy.
Pulse Analysis
The networking market has been racing to embed artificial intelligence after surveys showed that more than 90 % of enterprise IT leaders trust AI‑powered networking and are ready to invest. Extreme Networks, a long‑time player in campus and data‑center switches, seized the moment at its Connect 26 conference by announcing Agent ONE, a purpose‑built AI stack that moves beyond the prompt‑driven copilots that dominate today’s toolset. By integrating reasoning, live context and procedural knowledge, Extreme aims to turn the network itself into an autonomous decision‑maker, promising faster remediation and lower operational cost.
Agent ONE is organized into four layers: an infrastructure layer for raw reasoning, an AI core that supplies real‑time network context, a skills layer that codifies standard operating procedures, and an agentic layer that executes actions under a governance overlay. The first product, Agent ONE Coworker, arrives in July 2026 and works side‑by‑side with IT staff, surfacing insights, auto‑creating tickets and even adjusting Wi‑Fi parameters without a human prompt. A later release, Agent ONE Operator, will run continuously, autonomously handling scheduled workflows and event‑driven tasks while learning from each outcome.
For enterprises, the shift from reactive AI assistants to embedded autonomous agents could compress incident‑resolution cycles from weeks to minutes, freeing scarce network engineers for strategic projects. The accompanying Extreme Exchange marketplace further differentiates the offering by allowing customers and partners to publish industry‑specific skills, from healthcare device prioritization to retail POS traffic shaping. Competitors such as Cisco and Juniper are still focused on AI‑enhanced dashboards, so Extreme’s bold architecture may pressure the market toward tighter AI‑network integration, though governance, security and change‑management concerns will need careful handling.
Extreme Connect 26: Agent ONE takes forward network AI
Comments
Want to join the conversation?
Loading comments...