Cisco AI Assistant for Managing Firewall Policies Is Now Available

Cisco AI Assistant for Managing Firewall Policies Is Now Available

Cisco AI Assistant is now available for and Cisco Defense Orchestrator

Managing firewall policies and locating relevant documentation can be daunting for firewall administrators. However, the AI Assistant integrated with the Cisco Defense Orchestrator (CDO) and the cloud-delivered Firewall Management Center simplifies these processes. With this powerful combination, administrators can effortlessly manage firewall devices, configure policies, and access reference materials whenever required, streamlining their workflow and boosting overall efficiency.

Prerequisites

Administrators need to ensure they have met the following prerequisites to use the AI Assistant:

User roles:
● CDO and cloud-delivered Firewall Management Center – Super Admin or Admin
● On-Prem FMC – Global Domain Admin

Upon successful login into your tenant, you will notice an AI Assistant button positioned in the top menu bar of the dashboard.

Click the AI Assistant button on the CDO or cloud-delivered Firewall Management Center home page to access the AI Assistant.
The Cisco AI Assistant interface contains the following components: Text Input Box, New Chat, Chat History, Expand View, and Feedback.

ai-assistant-interface

Cisco AI Assistant interface following the best Generative AI assistant practices.

AI Assistant interaction

AI Assistant completion with the prompt “Can you provide me with the distinct IP addresses that are currently blocked by our firewall policies?”

blocked-addresses

AI Assistant completion with the prompt “What access control rules are disabled?”

ai disabled-rules

If you think that response is wrong, please click the thumbs-down button below for the related completion and fill out and submit the form.

ai feedback

AI Assistant can’t proceed with some prompts and questions. In this case, you can see the following completion:

ai default-completion

It looks like the engineering team decided not to display answers if there is insufficient data to correct them or in cases where the model can hallucinate.

Try it!

 

Share: