AI has to be auditable to be reliable

In this clip from an AI roundtable with Gremlin, Nobl9, and PagerDuty, Mandi Walls talks about how companies will want to audit AI to keep it reliable. Find out how else to improve AI reliability with Gremlin → https://www.gremlin.com/solutions/improve-ai-reliability

FULL TRANSCRIPT:
One of the things that I think we're gonna see, especially from enterprise customers, is they want auditing of what the agents are doing.

Where did they come from? What kicked it off? What events spurred this thing into action? What did it do? Where did it do it? What environment was it in? Any of those kinds of things. I think those expectations, what you have for basic automation don't change when there's agentic AI applied to that.

And I think that is gonna have to be part of what comes into some of this stuff is making sure that you're logging and auditing the actions taken by any of these components that are reaching out into your production systems. And that hopefully, it's something that you already have when you've got humans doing that stuff.

And it should just be naturally part of, "Yes, when we touch production, we are logging, we are making notes of what we're doing there."