As enterprises integrate greater levels of AI into their operational workflows, many are concerned about the effectiveness and accuracy of the decisions with which it's being entrusted. Mike Fratto joins host Eric Hanselman to talk about his recent research into the ways in which observability can provide greater context to AI systems and administrative teams. Data is the raw material from which AI value is created and a lack of solid operational data will mean that management systems won't have a complete perspective. Observability platforms have been expanding to collect and correlate a broader range of data and they can be particularly useful, if they're integrated well.

Observability data can also assist in building confidence in automation and support automation efforts by creating scripts and validating control decisions. It can be a huge help to staff to digest and make sense of existing run books and processes and guide decisions in the event of failures. When novel incidents occur, being able to rule out certain areas can be just as powerful as finding the root cause. No, the network isn't always the cause of the outage…

More S&P Global Content: Next in Tech | Ep. 271: AI Networking Next in Tech | Ep. 247: Security and Observability

For S&P Global subscribers: Updating processes and building trust are key to full operational autonomy – Highlights from VotE: DevOps Observability Market Monitor & Forecast Agent orchestration and observability go hand in hand: The case for governance in motion

Host/Author: Eric Hanselman

Guest: Mike Fratto

Producer/Editor: Dylan Scheible

Published With Assistance From: Feranmi Adeoshun and Sophie Carr