Grafana learns from Datadog's AI agent security lessons

Datadog published a list of lessons from securing AI agents — the kind of infrastructure that runs autonomously, calls other systems, and makes decisions without a human in the loop. The post is a field guide, not a sales pitch: it covers what went wrong when agents overreached, how to constrain them so they can't rewrite production configs, and how to audit their actions afterward. Grafana read it and built monitoring for HCP Terraform and Terraform Enterprise on top of Grafana Cloud.

The real story isn't the tools — it's the shift. AI agents are no longer a prototype in a lab. They're touching production infrastructure, and the security model is still catching up. The old playbook for monitoring machines and containers doesn't cover agents that think for themselves. That gap is where Grafana is putting its money.

Why this matters for us: if you're running agents that touch your infra, this is exactly the kind of monitoring you need before they start making mistakes in production.