Cisco Agentic AI Autonomously Resolves 145,000 Support Cases

What happened

Cisco has disclosed that its agentic AI systems have autonomously resolved 145,000 customer support cases, marking one of the more concrete data points yet on what autonomous AI agents can deliver in enterprise service operations. The company has framed the technology as serving a dual purpose: strengthening service delivery while also supporting commercial objectives such as customer retention.

The disclosure positions Cisco among a small group of large technology vendors now reporting hard resolution figures for agentic AI, rather than pilot programmes or roadmap commitments, signalling a shift from experimentation toward operational deployment at scale.

Why it matters

For CX leaders, this is a meaningful data point in an area still short on verified outcomes. Agentic AI has been widely discussed as a route to faster, cheaper support, but public, quantified proof of autonomous case resolution at this volume has been rare. Cisco's figure gives other operators a benchmark to test their own assumptions against when building the business case for autonomous support.

The framing also matters from a behavioural standpoint: resolution speed is only valuable to customers and to retention if the experience feels trustworthy and appropriately escalated when it should be. How Cisco defines "resolved" — and whether customers perceive these interactions as satisfactory, not just closed — will determine whether this becomes a genuine service-design advance or simply a faster way to close tickets.

By the numbers

• 145,000 support cases resolved autonomously by Cisco's agentic AI systems.

The Renascence take

Headline resolution numbers are easy to publish and hard to interrogate. The figure that actually matters to customers and to the business sits one layer below the one being reported.

A case count tells you volume, not value. The real test of agentic AI in support isn't how many tickets it closes autonomously, but whether those closures hold up — fewer repeat contacts, higher satisfaction on resolved issues, and clean handoffs when the AI reaches its limits. Operators chasing similar deployments should instrument for reopens, sentiment and escalation quality from day one, not just throughput, or they risk optimising for a metric that looks impressive in a press release but says little about whether customers actually felt served.