Reports

Cisco DevNet launched the "Month of Smart Connections" campaign in February 2026, focusing on building meaningful connections across the multi-vendor networking ecosystem through open standards and automation tooling. The campaign delivered a practical roadmap via a four-episode video series, each paired with its own code repository. The series covered automation workflows from foundational to advanced levels. Episode 1 highlighted using Cisco Crosswork Network Services Orchestrator (NSO) with a single RESTCONF API to pull and commit configurations across multi-vendor environments, alongside GitHub Copilot agents to help discover correct configuration URIs. Episode 2 explored various automation patterns including Python scripting, Ansible playbooks, and CI/CD pipelines, all built upon the vendor-agnostic OpenConfig data model and the gNMI protocol. Episode 3 focused on automation trustworthiness, demonstrating a customized testing framework for networking based on the open-source Robot Framework, leveraging OpenConfig and gNMI. Episode 4 showcased cutting-edge AI-driven automation. It built an AI agent using the open-source LibreChat platform, integrating multiple Model Context Protocol (MCP) servers (including pyATS, NetBox, GitHub, Draw.io) orchestrated via Docker Compose. This unified ChatOps interface enabled inventory synchronization, device audits, report generation and commit to GitHub, automatic issue creation, safeguarded configuration validation and deployment, and on-demand network diagram rendering. **Commentary**: This series systematically combines open standards, automation frameworks, testing methodologies, and AI agent workflows, clearly outlining the evolution of network automation from tool integration to intelligent collaboration. For enterprises grappling with multi-vendor complexity, the "open standards + concrete toolchain" approach offers significant reference value. Particularly, the AI agent architecture based on MCP provides a viable technical prototype for intent-driven and natural language interaction in network operations.

Cisco has released an IDC analyst brief highlighting the critical role of intelligent technical support in the AI era for business success. The brief indicates that the key differentiator for enterprise IT operations is shifting from infrastructure hardware upgrades to how operations are managed with intelligence and foresight. Cisco's "Cisco Support powered by Cisco IQ" solution is its modern AI-powered digital interface designed to address this shift. The solution offers five key capabilities: 1) **Proactive and Predictive Management**: Using AI to anticipate and resolve issues, reportedly reducing downtime by up to 50%. 2) **Operational Simplicity through Automation**: Automating routine workflows, reportedly reducing time to resolution by 60% to 90%. 3) **Unified Portal Experience**: Delivered via Cisco IQ with an API-first design and support for the Model Context Protocol (MCP) and Agent-to-Agent (A2A) communication for simplified integration. 4) **Skills Enhancement**: Providing AI-driven training paths. 5) **Security, Resilience, and Agility**: Combining automation and predictive analytics, with Cisco reporting that internally, cases resolved via end-to-end AI automation are up to 4 times faster. The brief emphasizes that enterprises are increasingly prioritizing vendor service quality, proactive support, and relationships. Leveraging its deep networking expertise, data from over 50 million networks, and proprietary AI, Cisco positions itself as a partner capable of delivering exclusive, differentiated intelligent support. **Commentary**: This brief is essentially Cisco's marketing promotion for its "Cisco IQ" intelligent support solution, using IDC industry insights to validate its product direction. Its key differentiator lies in the combination of Cisco's vast network ecosystem data and proprietary AI models. For enterprise IT decision-makers, this highlights the need to focus on the intelligence, proactivity, and integration capabilities of a vendor's support services, not just product features.

Cisco's security blog reviewed the advanced malware framework VoidLink disclosed in late 2025, using it as a case study to expose new blind spots in modern enterprise security. VoidLink is a cloud-native, Kubernetes-aware implant framework specifically designed for Linux cloud and container environments. Its technical features include: the ability to detect whether it's running on AWS, GCP, Azure, or inside Docker containers/Kubernetes Pods and adapt its behavior accordingly; employing fileless, stealth persistence techniques; and possessing a "compile-on-demand" capability that lays the foundation for AI-enabled attack frameworks, usable for establishing C2 infrastructure and internal reconnaissance. Cisco Talos analysis found this framework has been leveraged by advanced threat actors in real campaigns targeting technology and financial organizations. The article points out that attacks represented by VoidLink, ShadowRay 2.0, and NVIDIAScape indicate adversaries are systematically targeting workloads (containers, pods, AI inference jobs) as the primary attack surface, exploiting traditional infrastructure flaws and AI supply chain weaknesses (e.g., poisoned model repositories). To counter such kernel-level threats, Cisco emphasizes the importance of eBPF-based runtime security. Its Hypershield solution, developed by Isovalent, deploys eBPF programs within the kernel to observe and enforce policies on process executions, syscalls, file access, and network activity in real-time. It natively understands Kubernetes identities, enabling effective detection and mitigation of malicious behaviors like VoidLink that evade user-space tools. **Comment**: This article is essentially a deep technical marketing piece for Cisco's eBPF kernel security technology (Hypershield). The core argument is that traditional signature-based Endpoint Detection and Response (EDR) and Cloud Security Posture Management (CSPM) have "visibility" blind spots against cloud-native, AI-accelerated targeted attacks, necessitating a shift to kernel-level runtime security. It is recommended that security teams treat Kubernetes and AI workloads as first-class security assets and evaluate the real-time visibility and protection capabilities of technologies like eBPF.

Cisco emphasizes in its blog that the global telecommunications industry faces critical security challenges, with outdated network equipment, unpatched systems, and insecure protocols being exploited, highlighting the urgency of building more resilient infrastructure. Cisco asserts that network security is determined by its weakest link. In response, Cisco announces a redoubled commitment to resilient infrastructure and introduces a series of concrete measures to elevate the security baseline of its devices. These include: changing default security settings, deprecating and ultimately removing insecure capabilities and protocols, and introducing new security features. Cisco plans a phased removal strategy spanning three feature releases, implementing steps of warnings, restrictions, and final removal to minimize disruption to existing operations. Through these changes, Cisco aims to reduce the attack surface, protect sensitive data, and provide better visibility into threat actor activities. Its core philosophy is to not only make securing networks easy but to make operating them insecurely "ridiculously difficult." This is positioned as a key commitment from Cisco in building a secure future for global mobile networks with "always-on" performance in the AI era. <b>Comment</b>Cisco's proposed security enhancement measures are systematic. The phased strategy for retiring insecure protocols and features, in particular, demonstrates a mindset of improving security from the product design source, setting an example for the industry. It is advisable to monitor its specific implementation timeline and its impact on the compatibility of existing customer networks.

At MWC 2026, Cisco outlined its vision to transform mobility infrastructure into strategic business platforms. The core is the launch of the "Cisco Mobility Services Platform," designed to deliver mobility as a scalable, secure service with strict data sovereignty and security controls. The platform integrates several innovations: "IoT as a Service" to simplify global IoT operations at scale, the "Cisco Programmable Core" enabling global-scale networks without traditional operational overhead, and the "Cisco Mobility Services Gateway" empowering controlled migration from 2G through 5G Advanced toward AI-native, 6G-ready architectures. Cisco emphasizes that its platform strategy aims to directly generate new revenue or materially reduce costs for operators. Through a partnership with NVIDIA, Cisco is pursuing a dual strategy of "AI for Wireless" and "Wireless for AI," aiming to embed intelligence into service platforms to optimize operations while ensuring mobile infrastructure can support distributed AI workloads with predictable performance, low latency, and reliability. Furthermore, Cisco is joining the Linux Foundation's Open Centralized Unit Distributed Unit project as a founding member to drive the development of interoperable, cloud-native core frameworks. **Commentary**: Cisco is shifting from providing network components to delivering end-to-end cloud-native platforms. The strategic core is to systematically reduce operational complexity and accelerate service innovation through platform consolidation and service-based architectures, helping operators achieve profitable growth in the AI era. Its collaboration with NVIDIA on edge AI network architecture and its push for open standards with industry bodies are key layouts for future distributed intelligent workloads.

Cisco has launched Nexus Dashboard Data Broker, an innovative product designed to simplify the processing of network monitoring traffic for data center and campus networks. The core function of this solution is to act as an intelligent packet broker, capable of ingesting traffic from SPAN or TAP sources, performing operations like filtering, header modification, and packet deduplication, and then redirecting precise traffic to monitoring and analysis tools such as IPS and SIEM solutions. The product architecture is based on Cisco Nexus 9000 Series Switches and requires the NX-OS Essentials license. Its key technological innovations include: 1) **Low-touch Configuration**: Natively integrates with Cisco ACI data center fabrics, Catalyst Center-based campus fabrics, and standalone switches, enabling automatic configuration via a single GUI and simplifying deployment in heterogeneous environments. 2) **Hardware-accelerated Deduplication**: Introduced in NX-OS 10.4(1)F, supports TCP/UDP protocols, and can be performed in-line on switches or out-of-line via dedicated switches, effectively reducing the load on analysis tools. 3) **Visibility of Encapsulated Traffic**: Supports stripping headers and labels from various encapsulation protocols including MPLS, VXLAN, GRE, Q-in-Q, and ERSPAN, providing true end-to-end traffic visibility. 4) **Resiliency and Redundancy**: Features failover mechanisms, supports symmetric hashing for load balancing, and can automatically program backup paths upon component failure, ensuring continuous packet visibility. Additionally, it can generate NetFlow or sFlow for SPAN/TAP traffic, ensuring compatibility with a wider range of analysis tools. The solution is currently deployable on Linux hosts and is planned for future integration into the unified Nexus Dashboard platform, with its Nexus Dashboard 4.2 release set to formally incorporate the Data Broker functionality. **Comment**: This move strengthens Cisco's position in the network observability domain. By addressing the performance bottlenecks and cost issues of monitoring tools caused by processing redundant or encapsulated traffic through hardware-accelerated deduplication and traffic preprocessing, this product provides a key component for building efficient and intelligent monitoring architectures. Enterprises constructing or upgrading their network monitoring systems, especially those with complex encapsulated traffic or hybrid (data center/campus) environments, should consider a thorough evaluation.

Cisco has published an in-depth analysis on its official blog regarding the evolution of SD-WAN architecture, proposing a new paradigm of "Beyond the Controller" with decentralized intelligent networks. The article points out that traditional centralized SD-WAN architectures suffer from a "latency of logic" bottleneck, where the closed-loop process from edge device detection to controller decision-making and configuration deployment is too slow to cope with the rapidly changing traffic patterns and link quality in modern enterprise networks. The core of Cisco's proposed new architecture lies in distributing intelligence from the central controller to the network edge, transforming each edge device into an autonomous agent capable of Perception, Decision, Action, and Learning. This architecture is built on three pillars: 1. **Autonomous Decision-Making at the Edge**: Devices act as independent intelligent agents, making real-time routing decisions based on an understanding of their local environment and the global topology (via advanced learning techniques), continuing to optimize traffic and enforce policies even when disconnected from the central controller. 2. **Learning Networks**: Moving away from rule-based systems with fixed thresholds to a machine learning framework driven by reward mechanisms aligned with business objectives. The network learns specific traffic patterns (e.g., morning/evening rushes) through trial and error, enabling proactive prediction and prevention of issues rather than reactive responses. 3. **Intent-Driven Networks**: Utilizing technologies like natural language processing as a translation layer to automatically convert high-level business intent (e.g., "POS transactions always have priority during business hours") into executable technical policies (e.g., traffic classification, dynamic bandwidth reservation, path selection), forming a continuous verification and self-correcting assurance loop. Using a retail chain with over 500 sites as an example, the article contrasts the response differences between traditional centralized solutions and distributed intelligent agent solutions during sales peaks, with the latter enabling second-level localized decision-making to automatically prioritize critical business. Cisco also proposes a practical three-phase evolutionary path: Augmented Intelligence (AI-assisted human operators), Bounded Autonomy (automation for known scenarios), and Full Distribution (future vision). **Commentary**: The decentralized intelligence architecture proposed by Cisco essentially deepens the integration of the AI Agent concept into network infrastructure, aiming to solve the fundamental latency issue of centralized management. Its core value lies in elevating network operations from "configuring the network" to "declaring business intent" and enabling continuous autonomous optimization. This represents a significant technical direction for SD-WAN and the broader networking field, moving from automation to autonomy. Enterprise network teams should monitor its phased evolution path and evaluate the feasibility of gradually introducing edge intelligence and intent-driven capabilities into their existing SD-WAN foundations.

Cisco detailed its strategy for building an internal learning culture and tools in the AI era within its FY25 Purpose Report. The report is based on a multimodal analysis revealing employee AI usage patterns: leadership is the biggest predictor of adoption, with employees twice as likely to use AI if their leader does; additionally, employees with less than one year or more than 20 years of tenure showed the highest AI usage. 91% of employees learn AI through trial and error, while high performers embed AI into daily workflows to boost efficiency. To systematically upskill all employees, Cisco launched its flagship learning program, "AI for Everyone." It offers persona-based (Users, Leaders, Enablers, Builders) bite-sized lessons and hands-on exercises, ranging from foundational skills to AI data analytics, shifting from broad training to focused, role-relevant content. Complementing this, a standalone training on data privacy, inclusive design, and human-in-the-loop decision-making was developed to foster ethical AI adoption. On the tooling front, Cisco IT officially launched its next-generation internal AI assistant, CIRCUIT. This tool securely integrates the latest AI models into a cohesive system, handling tasks from drafting emails to unlimited deep research. CIRCUIT is designed to evolve, tackling more complex data, introducing intelligent agents, and laying the groundwork for connected, automated workflows—all with Cisco-grade trust and security at its core. **Comment**: By acting as "customer zero" for its own AI practices, Cisco has created a closed-loop strategy encompassing culture, training, and tools. This is not merely an internal efficiency solution but a validation ground for its "AI-first" technology philosophy, accumulating invaluable experience for future delivery of mature AI solutions and best practices to enterprise customers.

German drive and automation technology manufacturer SEW has deployed Cisco's Ultra-Reliable Wireless Backhaul (URWB) technology in its smart factory in Brumath, France, to address the reliable connectivity challenges of Autonomous Guided Vehicles (AGVs). Traditional Wi-Fi can cause communication interruptions during AGV handovers between different access point coverage zones, forcing AGVs to stop, impacting production safety and efficiency, and being susceptible to RF interference from metal structures. Cisco URWB provides an innovative solution. Its core advantage lies in fundamentally different roaming management: when an AGV moves to a new coverage zone, it establishes a new connection before breaking the old one, enabling seamless and uninterrupted handovers. Furthermore, URWB employs Multi-Path Operations (MPO) technology, ensuring reliability through spatial, frequency, and time diversity. High-priority packets can be duplicated up to eight times and sent via uncorrelated redundant paths and frequencies multiple times within milliseconds, guaranteeing command delivery even in the presence of interference and obstacles. This solution delivers significant benefits:它不仅为移动设备提供了比Wi-Fi更可靠的连接，还将所需接入点数量减少了约50%，降低了部署复杂度与成本。 **Comment**: Cisco URWB, with its unique "make-before-break" roaming and MPO technology, precisely meets the stringent requirements for determinism and reliability in industrial mobile scenarios. It is not merely a supplement to Wi-Fi but a dedicated wireless solution for mission-critical mobile applications like AGVs and AMRs, offering a new direction for the evolution of industrial wireless networks. This is noteworthy for the manufacturing and logistics automation sectors.

In January 2026, the German government launched a decade-long "Transformation Fund" (2026-2035), allocating up to €50 billion to drive the digital and structural modernization of its healthcare system. This ambitious initiative addresses systemic pressures from demographic aging and workforce shortages, with a projected shortfall of up to 1.9 million nursing professionals by 2040. At the core of the fund is the development of digital infrastructure, highlighted by the nationwide rollout of the electronic patient record (ePA). By the end of 2025, all insured persons became eligible for an ePA, with healthcare providers legally mandated to use compatible software and populate it with structured data. This creates a secure, standardized platform for data exchange, aligning with the European Health Data Space (EHDS). The transformation strategy focuses on six digital priorities: 1) Interoperable data platforms using open standards like FHIR to eliminate data silos; 2) Integration of telemedicine and networked care; 3) Adoption of flexible, service-based IT models; 4) Migration to the cloud under strict data privacy frameworks; 5) Enhanced cybersecurity and operational resilience; and 6) Application of AI in healthcare delivery to reduce administrative burdens and aid diagnostics. The plan also emphasizes workforce empowerment through skills training (e.g., Cisco Networking Academy) and public-private partnerships to ensure successful implementation.

Cisco demonstrated a deployable reference architecture for a “secure AI data platform” at VAST Forward 2026. Built on the design principles of “Cisco Secure AI Factory with NVIDIA,” the platform treats AI as an end-to-end system, unifying the design of performance, data readiness, cloud-native operations, and security. The platform integrates NVIDIA accelerated computing infrastructure and software, the VAST Data Platform, Cisco infrastructure, the Isovalent Enterprise Platform (based on Cilium and Tetragon), and Cisco AI Defense. The VAST Data Platform transforms enterprise data into usable context for AI workflows. On the security front, the platform employs a multi-layered approach to address AI-specific risks: The Isovalent Enterprise Platform based on Cilium controls east-west traffic within Kubernetes clusters using identity-based policies. Runtime security based on Tetragon provides kernel-level visibility into container behavior. Cisco AI Defense offers protection directly at the AI application layer, analyzing model vulnerabilities and applying runtime guardrails for prompts and responses. This reference architecture demonstrates full lifecycle security protection spanning data pipelines, infrastructure, Kubernetes operations, and AI application behavior. It aims to enable enterprises to deploy production-scale AI services safely and confidently, bridging the gap between innovation and trust. **Comment**: This solution marks a critical step in shifting AI deployment from model innovation to operational reality. Its core value lies in providing an end-to-end, actionable blueprint that integrates top-tier vendor technologies, particularly by deeply fusing traditional infrastructure security (K8s networking & runtime) with emerging AI-native security (model & agent protection). It offers a systematic approach to solving the security and governance challenges of scaling AI deployments.

Cisco's new joint report with Omdia, "The Race to Agentic AI: Why Infrastructure Will Make or Break Workforce Transformation," positions agentic AI as a central strategic imperative. Based on a survey of 650 executives across six countries, the report finds that 80% believe their company's survival will depend on agentic AI by 2027. Technically, it highlights a significant gap between existing infrastructure and the requirements for running agentic AI securely at scale, underscoring the urgent need for more robust networks. Executives predict that 55% of their workforce will collaborate with AI agents within the next 24 months, though legacy systems and a widening skills gap pose major hurdles. The research shows 87% of leaders have already reshaped strategic priorities to support agentic AI. In terms of investment, early adopters are allocating 37% of their technology budget to agentic AI, with 43% of these pioneers reporting meaningful ROI. Furthermore, 65% of organizations expect agentic AI to create entirely new job categories in 3-5 years, while nearly 60% of employees will need upskilling to supervise and trust autonomous agents. **Commentary**: The report elevates agentic AI from a "promising technology" to a "survival necessity." The core challenge lies in the synchronous transformation of infrastructure (network, data, security) and workforce skills. Companies should prioritize assessing and upgrading their network and security architecture while concurrently planning for employee reskilling to capture this transformation window.

At the recent Cisco Live EMEA and AI Summit in Amsterdam, Cisco positioned security as an accelerator for partner growth in the AI era and announced new solutions targeting "Agentic AI" security. Cisco asserts that as organizations transition from generative AI to autonomous "Agentic AI," security risks (such as prompt injection, tool poisoning, and data leakage) surge, presenting a significant market opportunity for partners. Cisco announced a major expansion of its security portfolio to protect the entire AI lifecycle. Key innovations include: 1) **AI Defense Expansion**: Now extended to protect the entire AI application supply chain and model integrity, integrated with the NVIDIA Secure AI Factory; 2) **SASE for Agentic AI**: Introduced automated detection and access controls for AI agents, including security for the Model Context Protocol; 3) **AgenticOps in Security Cloud Control**: Using AI to manage AI by analyzing firewall traffic and configuration data to autonomously remediate issues, reducing manual work for customers' IT teams. Additionally, Cisco launched **Duo for Active Directory** to counter identity attacks targeting legacy systems. To incentivize partners to specialize in AI security, Cisco officially launched the **Cisco 360 Partner Program** on January 25. The program features a "Specialization Bonus" mechanism, granting permanent extra bonuses to Preferred Partners with certifications in areas like "Secure Networking" and "Secure AI Infrastructure," encouraging deep technical expertise development. **Comment**: Cisco is deeply integrating security into the network fabric to address novel threats from Agentic AI. Its new solutions proactively embed and automate security capabilities, aiming to lower the barrier for AI adoption at scale. The new partner program directly incentivizes channel sales of its AI security solutions, demonstrating a clear strategy to rapidly capture this emerging market. It is advisable to monitor the progress of its collaboration with NVIDIA on AI factory security.

Cisco, in partnership with Australian neocloud leader Sharon AI, has launched Australia's first "Cisco Secure AI Factory with NVIDIA." This solution provides secure, scalable, and high-performance sovereign AI capabilities, ensuring all data and AI processing remain within Australia, aligning with the country's National AI Plan. The core technical architecture of this AI Factory consists of Cisco's UCS servers, its security and networking portfolio, and Nexus Hyperfabric technology (as part of the unified management plane, Nexus One). Its compute power is delivered by **1,024 NVIDIA Blackwell Ultra GPUs**, designed for complex AI and machine learning tasks. The solution also integrates VAST Data's cutting-edge storage systems for fast, reliable, and scalable data access, with all critical infrastructure hosted in NEXTDC's world-class Australian data centers. This initiative aims to accelerate enterprise AI adoption and innovation across the Asia-Pacific region. By strengthening sovereign capabilities and building a trustworthy AI ecosystem, it empowers enterprises and governments to leverage their data for differentiation and competitive advantage. Sharon AI will offer tailored solutions and a sandbox environment for various industries based on this platform. **Commentary**: This deployment represents a significant move by Cisco in the sovereign AI infrastructure space. By integrating its own networking/security hardware with NVIDIA's top-tier AI compute, Cisco has built a complete, closed-loop solution from hardware to management. Its "data-in-country" positioning precisely addresses the growing global regulatory focus on data security and AI sovereignty. This serves as a benchmark for Cisco to replicate similar solutions in the APAC region and globally, warranting close observation of its subsequent customer adoption and industry replication model.

Cisco and AT&T have announced a deepened strategic collaboration to jointly launch and commercially activate a 5G Standalone (SA)-native IoT platform. This platform tightly integrates AT&T's nationwide 5G SA core network with Cisco's leading Mobility Services Platform portfolio, including IoT Control Center and Converged Core offerings. The collaboration aims to provide end-to-end solutions for the most demanding IoT applications, ranging from connected vehicles and smart cities to digital healthcare. Its core value lies in unlocking programmable network capabilities, particularly network slicing and application-aware performance at scale, thereby delivering ultra-low latency, enhanced security, and exceptional reliability to enterprise customers. Furthermore, the platform plans to leverage Cisco's Converged Core and global distributed data centers (subject to regulatory constraints) for local breakouts to optimize performance for businesses operating in various global locations. Through Cisco's IoT Control Center, enterprises gain intuitive, unified tools for lifecycle management, diagnostics, and automation, simplifying the deployment and management of IoT solutions. **Comment**: This collaboration signifies a deep integration between a telecom operator and an equipment provider, jointly offering a "network + platform" integrated solution for the enterprise 5G IoT market. Its key differentiation lies in the native 5G SA architecture's support for critical features like network slicing, directly targeting industry applications with extremely high demands for performance and security. It is advisable to monitor the platform's subsequent specific industry deployment cases and its integration with edge computing and AI-driven IoT dynamic adaptation capabilities.