GlossaryProductsIntermediate

OpenTelemetry

Also written: OTel, OpenTelemetry profiling, OTel profiling

An open-source framework for collecting and exporting data about software performance and behavior.

In plain terms

OpenTelemetry is a way to track what software is doing. Think of it like a dashboard for a complex machine. It collects information like how fast a process runs, if it encounters errors, and the sequence of operations.

This data helps developers understand if their software is working correctly and efficiently. For example, if an AI chatbot is slow to respond, OpenTelemetry can show which part of the system is causing the delay, helping engineers fix it.

The expert version

OpenTelemetry is an open-source observability framework designed to standardize the generation, collection, and export of telemetry data. It provides APIs, SDKs, and tools for instrumenting applications to produce metrics, logs, and traces. This unified approach allows developers to gain insights into the performance and behavior of their systems, regardless of the underlying infrastructure or programming language.

It aims to simplify observability by offering vendor-neutral instrumentation. Data collected via OpenTelemetry can be sent to various backends for analysis and visualization, such as Jaeger for distributed tracing or Prometheus for metrics.

Why it matters now

OpenTelemetry is appearing in frontier AI news as companies like Anthropic, Google, and Microsoft use it to monitor and improve their AI systems. This open-source framework helps track the performance and behavior of AI agents and applications, allowing developers to catch bugs and understand how their tools are interacting in production environments.

Where it shows up 6 items, newest first
How each company uses it
Anthropic Claude
Anthropic is using OpenTelemetry to help developers track their Claude API calls.
Google Gemini
Google is using OpenTelemetry to help automate AI agent testing and catch production bugs.
Microsoft Copilot
Microsoft Copilot uses OpenTelemetry for client telemetry and to gain deeper insight into its AI agent performance and tool interactions, especially when integrated with tools like Microsoft Aspire.
Related terms

ObservabilityDistributed TracingMetricsLoggingAPMTelemetry

Often confused with
OpenTracing
An older, now-deprecated project that OpenTelemetry absorbed. OpenT
OpenCensus
Another observability project that merged with OpenTracing to form O
Prometheus
A popular monitoring and alerting system that can ingest data collec
Origin

OpenTelemetry was created by the CNCF (Cloud Native Computing Foundation) as a merger of two previous projects: OpenTracing and OpenCensus. The goal was to create a single, vendor-neutral standard for observability. Source ↗

Try this about 5 minutes

You can explore OpenTelemetry by looking at the documentation for your favorite programming language. Many web frameworks have libraries that make it easy to add OpenTelemetry instrumentation. For example, you could try adding basic tracing to a simple web application to see how requests are tracked.

Try it: Playbook entries that use thisPlaybook →

The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.

Questions people ask

What kind of data does OpenTelemetry collect?
OpenTelemetry collects metrics (numerical measurements over time), logs (timestamped records of events), and traces (the end-to-end journey of a request through a system).
Is OpenTelemetry tied to a specific cloud provider?
No, OpenTelemetry is vendor-neutral. It is designed to work with any cloud provider or on-premises infrastructure, sending data to various analysis tools.

First seen on Subvolts Sep 7, 2026. Updated Oct 14, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.