1+1=3 ... Agentic AI and GenAI Reduce Barriers to Network Automation in Network Operations

Song Pang
NetBrain Technologies

Artificial intelligence is transforming network operations (NetOps), supercharging automation, enabling new predictive capabilities, improving visibility and powering nearly continuous optimization. Two flavors of AI — generative AI and agentic AI — bring different but complementary capabilities to the latest generation of network automation platforms, with generative AI helping users to digest and understand what's happening in the network, while agentic AI enables greater automation of common tasks and faster (even autonomous) responses to anomalies.

When trained on live network data and large libraries of validated automations, GenAI and agentic AI can help NetOps teams be more productive while making networks more reliable, easier to manage and secure. And as agentic AI becomes more powerful, AI's overall usefulness to the NetOps team will increase exponentially.

Generative and Agentic AI - Improving NetOps in Different Ways

In NetOps, the volume and velocity of data can at times seem overwhelming. Generative AI excels at creating new content from existing data. By reading and interpreting raw data and turning it into human-readable content and insights, GenAI can help NetOps teams to improve their visibility of network status and streamline problem identification and resolution. Some specific use cases include:

  • Processing CLI command results: GenAI can take the raw output of CLI commands (CLI results, logs) and interpret them. For instance, if the CLI output contains a list of errors, GenAI can be used to analyze the results, correlate them with known issues or network states, and summarize the findings in a more human-understandable format.
  • Generating answers: It isn't always easy for humans to extract knowledge from data. But GenAI makes it much easier for NetOps teams (and other users) to query network data and get answers to questions or actionable insights.
  • Explaining automation results: Many NetOps teams automate tasks like firmware updates or password changes using Ansible, homegrown tools, or commercial platforms. GenAI can read and interpret the results of these automations when they're run at scale. For example, if NetOps ran an automation to update 100 Cisco routers, GenAI can check the results of each and report that 94 were successful, 4 failed due to one issue, and two failed because of a different issue.

Agentic AI, by contrast, is designed to enable autonomous, goal-directed behavior, including making decisions, based on predefined rules, triggers, or specific requests. In the case of NetOps, agentic AI can be used for:

  • Running CLI Commands: For example, agentic AI could execute CLI commands on network devices or automation tools, checking the status of routers, switches, or firewalls and adjusting configurations as needed.
  • Taking follow-up actions: After the execution of automated scripts or commands, Agentic AI can be used to read the results, understand the status, and take appropriate follow-up actions. For example, if a script encountered errors, an AI agent could re-run the task, alert administrators, or log the issue for review.
  • Creating dashboards: Agentic AI can dynamically generate and update observability dashboards based on real-time network data. This can include visualizing performance, fault detection, or resource usage, and updating the view based on the latest metrics, thresholds, or events.
  • Improving device visibility: AI agents can retrieve device basic properties, such as IP addresses, L2 & L3 neighbor, etc.

New Technology, New Challenges

While the existing and potential benefits of AI for NetOps are growing every day, the reality is that the integration of this powerful technology is still in its infancy. It's important to understand the specific challenges and limitations of this technology and set up systems and processes to manage them.

First, agentic AI requires high levels of trust and reliability. Businesses depend on their networks and outages are extremely expensive. AI must demonstrate that it can consistently make accurate diagnoses and execute effective remediation steps before network teams will be comfortable ceding a degree of autonomy to the system. For network automation platform vendors, this means rigorous testing and validation of the agentic AI models across a wide range of network scenarios and conditions. For the teams or individuals that adopt a new AI tool, this means using PoCs or a phased rollout to demonstrate that reliability and slowly win over other engineers and teams.

Another substantial challenge lies in integration with existing systems. Building AI features into existing network management tools and ensuring they are compatible with diverse network environments and a multitude of vendors is a complex technical undertaking. We shouldn't underestimate the amount of work it will take to realize the benefits of AI.

Furthermore, agentic AI models are inherently data intensive. To be successful, they need to access a large amount of high-quality network data, including configuration information, performance metrics, logs, and historical incident data. For agentic AI systems to become successful, NetOps teams may need to create new data collection mechanisms, improve data storage and processing capabilities, and develop data governance strategies.

As with any AI system, the usefulness and accuracy of the output is dependent on the quality of the training data it's built on. The ideal model for AI in NetOps is learning network know-how from human engineers; the larger the library of network automations that the AI can reference, the better it will be able to reason. But AI gets much more useful if it can use data from these automations, as well as live network data (from a digital twin) to reason or for context. All these features together add up to greater than the sum of their parts.

Song Pang is SVP of Engineering at NetBrain Technologies

Related Links

Hot Topics

The Latest

OpenTelemetry: A Complete Telemetry Framework that Grows with You

Most organizations approach OpenTelemetry as a collection of individual tools they need to assemble from scratch. This view misses the bigger picture. OpenTelemetry is a complete telemetry framework with composable components that address specific problems at different stages of organizational maturity. You start with what you need today and adopt additional pieces as your observability practices evolve ...

The Quiet Power of Milliseconds in High Traffic Systems

One of the earliest lessons I learned from architecting throughput-heavy services is that simplicity wins repeatedly: fewer moving parts, loosely coupled execution (fewer synchronous calls), and precise timing metering. You want data and decisions to travel the shortest possible path. The goal is to build a system where every strategy and each line of code (contention is the key metric) complements the decision trees ...

Are AI "Autonomous Coworkers" Ready for Payroll?

As discussions around AI "autonomous coworkers" accelerate, many industry projections assume that agents will soon operate alongside human staff in making decisions, taking actions, and managing tasks with minimal oversight. But a growing number of critics (including some of the developers building these systems) argue that the industry still has a long way to go to be able to treat AI agents like fully trusted teammates ...

Enterprise AI Reaches Pivotal Inflection Point: From Automation to Agentic Intelligence

Enterprise AI has entered a transformational phase where, according to Digitate's recently released survey, Agentic AI and the Future of Enterprise IT, companies are moving beyond traditional automation toward Agentic AI systems designed to reason, adapt, and collaborate alongside human teams ...

The Real Reason Enterprise AI Stalls: It Doesn't Fit in the Box

The numbers back this urgency up. A recent Zapier survey shows that 92% of enterprises now treat AI as a top priority. Leaders want it, and teams are clamoring for it. But if you look closer at the operations of these companies, you see a different picture. The rollout is slow. The results are often delayed. There's a disconnect between what leaders want and what their technical infrastructure can handle ...

AI's Inflection Point Will Redefine Enterprise Readiness in 2026

Kyndryl's 2025 Readiness Report revealed that 61% of global business and technology leaders report increasing pressure from boards and regulators to prove AI's ROI. As the technology evolves and expectations continue to rise, leaders are compelled to generate and prove impact before scaling further. This will lead to a decisive turning point in 2026 ...

Turning Foresight into Resilience: Reclaiming Prevention in the Age of Exposure

Cloudflare's disruption illustrates how quickly a single provider's issue cascades into widespread exposure. Many organizations don't fully realize how tightly their systems are coupled to thirdparty services, or how quickly availability and security concerns align when those services falter ... You can't avoid these dependencies, but you can understand them ...

MLOps Meets DataOps: Creating Unified Backbone for AI

If you work with AI, you know this story. A model performs during testing, looks great in early reviews, works perfectly in production and then slowly loses relevance after operating for a while. Everything on the surface looks perfect — pipelines are running, predictions or recommendations are error-free, data quality checks show green; yet outcomes don't meet the ground reality. This pattern often repeats across enterprise AI programs. Take for example, a mid-sized retail banking and wealth-management firm with heavy investments in AI-powered risk analytics, fraud detection and personalized credit-decisioning systems. The model worked well for a while, but transactions increased, so did false positives by 18% ...

2026 Network Monitoring Trends

Basic uptime is no longer the gold standard. By 2026, network monitoring must do more than report status, it must explain performance in a hybrid-first world. Networks are no longer just static support systems; they are agile, distributed architectures that sit at the very heart of the customer experience and the business outcomes ... The following five trends represent the new standard for network health, providing a blueprint for teams to move from reactive troubleshooting to a proactive, integrated future ...

2026 AI Predictions - Part 5

APMdigest's Predictions Series concludes with 2026 AI Predictions — industry experts offer predictions on how AI and related technologies will evolve and impact business in 2026. Part 5, the final installment, covers AI's impacts on IT teams ...

1+1=3 ... Agentic AI and GenAI Reduce Barriers to Network Automation in Network Operations

Song Pang
NetBrain Technologies

Artificial intelligence is transforming network operations (NetOps), supercharging automation, enabling new predictive capabilities, improving visibility and powering nearly continuous optimization. Two flavors of AI — generative AI and agentic AI — bring different but complementary capabilities to the latest generation of network automation platforms, with generative AI helping users to digest and understand what's happening in the network, while agentic AI enables greater automation of common tasks and faster (even autonomous) responses to anomalies.

When trained on live network data and large libraries of validated automations, GenAI and agentic AI can help NetOps teams be more productive while making networks more reliable, easier to manage and secure. And as agentic AI becomes more powerful, AI's overall usefulness to the NetOps team will increase exponentially.

Generative and Agentic AI - Improving NetOps in Different Ways

In NetOps, the volume and velocity of data can at times seem overwhelming. Generative AI excels at creating new content from existing data. By reading and interpreting raw data and turning it into human-readable content and insights, GenAI can help NetOps teams to improve their visibility of network status and streamline problem identification and resolution. Some specific use cases include:

  • Processing CLI command results: GenAI can take the raw output of CLI commands (CLI results, logs) and interpret them. For instance, if the CLI output contains a list of errors, GenAI can be used to analyze the results, correlate them with known issues or network states, and summarize the findings in a more human-understandable format.
  • Generating answers: It isn't always easy for humans to extract knowledge from data. But GenAI makes it much easier for NetOps teams (and other users) to query network data and get answers to questions or actionable insights.
  • Explaining automation results: Many NetOps teams automate tasks like firmware updates or password changes using Ansible, homegrown tools, or commercial platforms. GenAI can read and interpret the results of these automations when they're run at scale. For example, if NetOps ran an automation to update 100 Cisco routers, GenAI can check the results of each and report that 94 were successful, 4 failed due to one issue, and two failed because of a different issue.

Agentic AI, by contrast, is designed to enable autonomous, goal-directed behavior, including making decisions, based on predefined rules, triggers, or specific requests. In the case of NetOps, agentic AI can be used for:

  • Running CLI Commands: For example, agentic AI could execute CLI commands on network devices or automation tools, checking the status of routers, switches, or firewalls and adjusting configurations as needed.
  • Taking follow-up actions: After the execution of automated scripts or commands, Agentic AI can be used to read the results, understand the status, and take appropriate follow-up actions. For example, if a script encountered errors, an AI agent could re-run the task, alert administrators, or log the issue for review.
  • Creating dashboards: Agentic AI can dynamically generate and update observability dashboards based on real-time network data. This can include visualizing performance, fault detection, or resource usage, and updating the view based on the latest metrics, thresholds, or events.
  • Improving device visibility: AI agents can retrieve device basic properties, such as IP addresses, L2 & L3 neighbor, etc.

New Technology, New Challenges

While the existing and potential benefits of AI for NetOps are growing every day, the reality is that the integration of this powerful technology is still in its infancy. It's important to understand the specific challenges and limitations of this technology and set up systems and processes to manage them.

First, agentic AI requires high levels of trust and reliability. Businesses depend on their networks and outages are extremely expensive. AI must demonstrate that it can consistently make accurate diagnoses and execute effective remediation steps before network teams will be comfortable ceding a degree of autonomy to the system. For network automation platform vendors, this means rigorous testing and validation of the agentic AI models across a wide range of network scenarios and conditions. For the teams or individuals that adopt a new AI tool, this means using PoCs or a phased rollout to demonstrate that reliability and slowly win over other engineers and teams.

Another substantial challenge lies in integration with existing systems. Building AI features into existing network management tools and ensuring they are compatible with diverse network environments and a multitude of vendors is a complex technical undertaking. We shouldn't underestimate the amount of work it will take to realize the benefits of AI.

Furthermore, agentic AI models are inherently data intensive. To be successful, they need to access a large amount of high-quality network data, including configuration information, performance metrics, logs, and historical incident data. For agentic AI systems to become successful, NetOps teams may need to create new data collection mechanisms, improve data storage and processing capabilities, and develop data governance strategies.

As with any AI system, the usefulness and accuracy of the output is dependent on the quality of the training data it's built on. The ideal model for AI in NetOps is learning network know-how from human engineers; the larger the library of network automations that the AI can reference, the better it will be able to reason. But AI gets much more useful if it can use data from these automations, as well as live network data (from a digital twin) to reason or for context. All these features together add up to greater than the sum of their parts.

Song Pang is SVP of Engineering at NetBrain Technologies

Related Links

Hot Topics

The Latest

OpenTelemetry: A Complete Telemetry Framework that Grows with You

Most organizations approach OpenTelemetry as a collection of individual tools they need to assemble from scratch. This view misses the bigger picture. OpenTelemetry is a complete telemetry framework with composable components that address specific problems at different stages of organizational maturity. You start with what you need today and adopt additional pieces as your observability practices evolve ...

The Quiet Power of Milliseconds in High Traffic Systems

One of the earliest lessons I learned from architecting throughput-heavy services is that simplicity wins repeatedly: fewer moving parts, loosely coupled execution (fewer synchronous calls), and precise timing metering. You want data and decisions to travel the shortest possible path. The goal is to build a system where every strategy and each line of code (contention is the key metric) complements the decision trees ...

Are AI "Autonomous Coworkers" Ready for Payroll?

As discussions around AI "autonomous coworkers" accelerate, many industry projections assume that agents will soon operate alongside human staff in making decisions, taking actions, and managing tasks with minimal oversight. But a growing number of critics (including some of the developers building these systems) argue that the industry still has a long way to go to be able to treat AI agents like fully trusted teammates ...

Enterprise AI Reaches Pivotal Inflection Point: From Automation to Agentic Intelligence

Enterprise AI has entered a transformational phase where, according to Digitate's recently released survey, Agentic AI and the Future of Enterprise IT, companies are moving beyond traditional automation toward Agentic AI systems designed to reason, adapt, and collaborate alongside human teams ...

The Real Reason Enterprise AI Stalls: It Doesn't Fit in the Box

The numbers back this urgency up. A recent Zapier survey shows that 92% of enterprises now treat AI as a top priority. Leaders want it, and teams are clamoring for it. But if you look closer at the operations of these companies, you see a different picture. The rollout is slow. The results are often delayed. There's a disconnect between what leaders want and what their technical infrastructure can handle ...

AI's Inflection Point Will Redefine Enterprise Readiness in 2026

Kyndryl's 2025 Readiness Report revealed that 61% of global business and technology leaders report increasing pressure from boards and regulators to prove AI's ROI. As the technology evolves and expectations continue to rise, leaders are compelled to generate and prove impact before scaling further. This will lead to a decisive turning point in 2026 ...

Turning Foresight into Resilience: Reclaiming Prevention in the Age of Exposure

Cloudflare's disruption illustrates how quickly a single provider's issue cascades into widespread exposure. Many organizations don't fully realize how tightly their systems are coupled to thirdparty services, or how quickly availability and security concerns align when those services falter ... You can't avoid these dependencies, but you can understand them ...

MLOps Meets DataOps: Creating Unified Backbone for AI

If you work with AI, you know this story. A model performs during testing, looks great in early reviews, works perfectly in production and then slowly loses relevance after operating for a while. Everything on the surface looks perfect — pipelines are running, predictions or recommendations are error-free, data quality checks show green; yet outcomes don't meet the ground reality. This pattern often repeats across enterprise AI programs. Take for example, a mid-sized retail banking and wealth-management firm with heavy investments in AI-powered risk analytics, fraud detection and personalized credit-decisioning systems. The model worked well for a while, but transactions increased, so did false positives by 18% ...

2026 Network Monitoring Trends

Basic uptime is no longer the gold standard. By 2026, network monitoring must do more than report status, it must explain performance in a hybrid-first world. Networks are no longer just static support systems; they are agile, distributed architectures that sit at the very heart of the customer experience and the business outcomes ... The following five trends represent the new standard for network health, providing a blueprint for teams to move from reactive troubleshooting to a proactive, integrated future ...

2026 AI Predictions - Part 5

APMdigest's Predictions Series concludes with 2026 AI Predictions — industry experts offer predictions on how AI and related technologies will evolve and impact business in 2026. Part 5, the final installment, covers AI's impacts on IT teams ...