Necessary approval should be secured for business-critical issues. You don't want to resolve one issue to in turn create a bigger one, and you do not want to restrict the capability of GenAI.
There are pros and cons with AI, choose wisely when it is required.
I’d be open to it—with proper guardrails and oversight. Automation is powerful, but accountability is key.
If the AI can apply the runbook appropriately given the detected incident - then let it do it. I would rather an automated approach, applied correctly, then let it loose! Too many slow or bad decisions or worse, performing the wrong steps manually in a crisis. Automation will save you... Resistance is futile.
\m/
One big cost behind all this AI is the energy needed by the datacenters. Mitigating this is a significant concern, for multiple reasons.
The right answer: A properly trained model could indeed make autonomous decisions; would not fit for every operational duty.
Some issues could be auto-remediated, but business-critical issues need human approval
LET SKYNET RUN FREE!
Cisco Firepower Threat Defense FTD-1.3.6.1.4.1.9.1.2320.ConfigMgmt-CommandsExported with NCM 2023.4.2
AlertSummary.xmlHi All, I've done some work to tidy up our alert definitions (both the Alert Manager and Alert Actions) and the more important alerts have now been tailored and given a "[ *** ]" prefix so it is easy to see the alerts we have catered for versus those that might appear once in a while. I am monitoring this…
I wanted to share a new dashboard a co-worker ( @mblackburn ) came up with to help us better view groups of interfaces and various metrics for each. Using a single HTML widget, we are able to visually compare numerous different interfaces by populating each interface dataset onto its own graph. We are able to view…
PALO-ALG - Logging.xmlPalo Alto Networks ALG Security Technical Implementation Guide :: Version 2, Release: 3 Benchmark Date: 27 Oct 2022
