Share with your CIO
Cisco is betting that agentic AI, meaning AI systems that take multi-step actions autonomously rather than just answering questions, can restructure how a $15 billion services organization actually runs. Liz Centoni, EVP and chief customer experience officer, is leading that effort across 20,000 employees covering professional services, customer success, technical support, and renewals. The Cisco agentic AI program runs in parallel with the company’s $28 billion Splunk integration and positions Cisco as an AI-native operator, not just an AI vendor.
What this means for your business
The story here isn’t really about Cisco’s four pillars. It’s about what happens when a 26-year company veteran with P&L ownership over a services operation the size of a mid-cap company decides that AI isn’t a productivity add-on but a structural redesign. If your services or customer success organization still treats AI as a tool layered onto existing workflows, Cisco’s posture suggests you’re already operating a generation behind the firms setting the new baseline.
Cisco has an obvious incentive to present its own transformation as proof that the networking and security infrastructure it sells is the right foundation for agentic AI, so the four-pillar framing almost certainly flatters the company’s existing product portfolio. That doesn’t make the underlying bet wrong. What it does mean is that any CIO reading this framework should pressure-test whether the pillars hold outside a Cisco-native environment. The recurring failure mode in enterprise AI transformation isn’t bad strategy at the top, it’s integration debt at the layer where agents actually need to act, pulling data from dozens of systems that weren’t designed to talk to each other.
The Splunk acquisition sharpens the read. Splunk’s core value is telemetry, the system data showing how infrastructure and applications are actually performing in real time. Agentic AI without high-quality, real-time telemetry is essentially an agent operating on stale maps. Cisco buying Splunk and then building an agentic AI layer on top of it isn’t coincidental. CIOs who don’t have an equivalent data foundation should treat this as the leading indicator worth watching, because the gap between organizations with clean operational data and those without it will determine who can actually deploy agents that do something useful.
Concept deep-dive: Agentic AI
Agentic AI refers to AI systems that don’t just respond to a single prompt but plan and execute sequences of actions to complete a goal, often calling tools, querying databases, or triggering workflows along the way. Think of it less like a search engine and more like a junior analyst who can be handed an objective and will figure out the steps. The business significance is that it moves AI from advisory output to operational action, which changes both the value ceiling and the risk surface.
Based on reporting from Cisco’s 4 pillars for achieving agentic AI transformation, originally published 2026-11-13 05:05:00.

