While many tech firms focus on AI models and algorithms, Cisco has spent early 2025 building out the practical infrastructure that makes AI work in real business environments.
Their recent launches across collaboration, networking, and security demonstrate an understanding that effective AI requires more than just advanced mathematics – it needs reliable systems to deliver tangible outcomes.
Reimagining Work and Customer Interactions
At Enterprise Connect in March, Cisco unveiled AI-powered tools designed to bridge the gap between theoretical AI capabilities and daily business challenges.
The Webex AI Agent (arriving March 31) handles customer inquiries and executes tasks without wait times, working alongside human agents rather than replacing them. For internal teams, new workflow automation in the AI Assistant for Webex Suite connects processes across enterprise applications like Salesforce and ServiceNow, addressing the fragmentation that often undermines productivity.
The new Webex Calling Customer Assist transforms how organizations handle customer support, enabling any employee to assist customers through their existing Webex app. This approach removes the traditional barriers between specialized support agents and the rest of the organization.
Jeetu Patel, Executive Vice President at Cisco, spoke to this subtle but profound shift in business operations:
Enterprises are discovering how AI reshapes collaboration between people and technology.
Cross-platform integration has also been prioritized: Apple AirPlay support for Cisco Devices in Microsoft Teams Rooms, Spatial Meetings for Apple Vision Pro, and connectors through Glean for enterprise applications reflect an understanding that AI must work within mixed technology environments.
For healthcare providers, a native Epic integration for Webex Contact Center (in beta) brings patient information directly into communication workflows.
Building AI’s Foundation
At Mobile World Congress, Cisco expanded its Agile Services Networking solution, recognizing that AI deployments require robust network infrastructure to succeed.
The architecture employs Silicon One routing and converged IP/optics, with telecom providers Arelion, Lumen, and Reliance Jio already implementing it.
New tools include ThousandEyes Connected Devices for visibility into user experiences and Provider Connectivity Assurance, which uses traffic analysis to optimize network performance.
Cisco’s partnership with NVIDIA has deepened, connecting Silicon One with NVIDIA SuperNICs and becoming the first partner silicon integrated into the Spectrum-X network. This collaboration addresses real-world deployment challenges like congestion management and load balancing; the unglamorous but essential aspects of AI infrastructure that determine whether systems perform as promised.
The development of a framework combining NVIDIA silicon with Cisco’s operating system shows an effort to improve back-end connectivity and industry standards, potentially simplifying what remains a complex integration landscape.
Securing AI in Practice
Recognizing the unique security challenges of enterprise AI, Cisco launched AI Defense and Hypershield in January, addressing vulnerabilities that traditional security tools miss.
AI Defense (releasing March) performs automated testing of AI models for hundreds of potential vulnerabilities while continuously monitoring for threats like prompt injection. It gives security teams visibility into AI tool usage across organizations, helping to prevent unauthorized access and data leakage.
Hypershield takes a novel approach to data center security with a rule engine that identifies and addresses vulnerabilities within minutes. Its self-adapting network segmentation evolves based on system behavior, while validating security updates against live traffic without performance impacts.
“The data center will grow substantially as AI deployments accelerate,” noted Tom Gillis of Cisco’s Security Business Group, highlighting the scale challenge that many organizations underestimate when planning AI initiatives.
Kent Noyes from World Wide Technology added:
The adoption of AI brings risks that traditional cybersecurity doesn’t address. Cisco AI Defense provides visibility and protection against these evolving threats.
What This Means for Enterprise Strategy
Cisco’s approach tackles three fundamental challenges that often determine success or failure in enterprise AI:
- Infrastructure readiness: Their networking solutions and NVIDIA partnership provide the physical and digital foundation for AI deployments.
- Security maturity: AI Defense and Hypershield protect AI systems throughout their lifecycle, addressing both known and emerging threats.
- Workflow integration: Webex tools connect AI capabilities with existing business processes, increasing the likelihood of adoption and value creation.
With only 13% of organizations fully prepared for AI deployment while 90% seek benefits within two years, this grounded approach offers a path forward without requiring wholesale transformation.
Organizations in healthcare, manufacturing, and service sectors are already seeing returns on targeted implementations. As AI moves from experimentation to operation, those who address the full technology stack – not just the algorithms – will likely extract more value from their investments.
