Heavy AI marketing language makes parts of the product look like a copyable ‘AI layer’, even though core telemetry is non-trivial.

Multiple signals of LLM integrations and mentions of OpenAI/GPU monitoring suggest meaningful reliance on external model ecosystems.

Agent install, alerts, notebooks, incident/case workflows and CI visibility indicate daily, mission-critical usage that's hard to dislodge.

Deep cloud vendor and platform integrations plus events and marketplace presence create healthy channel and ecosystem entrenchment.

Wide telemetry coverage (metrics, traces, logs, profiles), platform integrations, GPU monitoring and on‑prem mentions point to genuine platform entanglement.

Compliance references, governance controls, analyst leader badges and explicit enterprise signals show procurement-friendly, durable trust posture.

Installed agents, historical telemetry, dashboards, and embedded incident workflows create significant data and collaboration gravity.

Clear enterprise GTM signals, analyst recognition, events and large customer claims show mature monetization, though pricing is only partially visible.

Infrastructure platforms start safer because they tend to sit deeper in the stack.

Small upward adjustment — Datadog remains deeply entrenched, but pervasive AI marketing and explicit LLM/GPU signals modestly increase copyability and model-dependency risk.

Cloud Monitoring as a Service | Datadog