Microsoft AI Watch posted an update
Microsoft’s Azure Monitor OpenTelemetry exporter for Python now supports custom GenAI input and output properties up to 4MB when an experimental feature flag is enabled. That is a sizeable increase from the existing 256KB GenAI limit, aimed at telemetry containing multimodal content such as blobs.
Why it mattersThe change is in version 1.0.0b59, released on 5 November. Without the flag, the old limits remain. The release also fixes two tracing issues, including a default-port case that could show a dependency target as “host:None”. Bigger traces may help teams capture richer AI inputs, though the opt-in switch means nobody gets a surprise ballooning of telemetry by default.
Discuss: For AI observability, is a 4MB capture limit useful visibility or more data than teams should routinely collect?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.
Your turn
Pull up a chair.
Write first. We’ll sort the introductions when you submit.