Live  /  AssemblyAI  /  Incident

AssemblyAI: Increased processing times and errors for Universal-3.5 Pro on the EU async endpoint

AssemblyAI · assemblyai.comMajor incidentResolved2026-10-05 11:39 UTC
Public — what happened

Summary

AssemblyAI reported a major incident starting at 10:21 UTC on 2026-10-05 in which asynchronous transcription jobs sent to its EU endpoint ran much slower than normal, with some requests failing outright. The company first named its Universal-3.5 Pro model as affected, later adding Universal-3 Pro, and said its US endpoint and real-time services were not impacted. AssemblyAI said it had pinpointed the problem by 10:38 UTC, though it did not describe what it was, and by 10:43 UTC processing times were returning to normal while it kept watching. The incident was marked resolved at 11:39.

Affected

Asynchronous API

As listed on AssemblyAI’s status page.

Timeline

Source of record: provider status page. Summary written by SquidGrid from the provider’s published updates; times in UTC.

Independent path health

The status page is the symptom. SquidGrid measures the paths that reach AssemblyAI against their own baselines — whether the routes carrying your traffic degraded, at which handoff, and who owns that stretch.

🔒 Sign in to see
Customers get the live Provider Assessment Report for AssemblyAI: its full incident track record, what tends to break, and recent public reports of problems.

Included with any paid plan.

Open the AssemblyAI report →

More from AssemblyAI

Covered in The Squid’s Ink, Week 42, 2026, our weekly internet health report.

The full archive is on the AssemblyAI status & history page.

Related: AssemblyAI status & history. SquidGrid is not affiliated with AssemblyAI.