Power Platform Community Forum Thread Details

Currently, the agent created in Copilot Studio is returning a system error and is not functioning.
Could you please help identify the cause and suggest a resolution?

[Error Message]
Bot says:
An unexpected error has occurred. We are currently investigating. Error code: SystemError. Conversation ID: 3c59ea9d-038f-4f59-885c-eb37665f257f. Time (UTC): 6/12/2026 8:26:56 AM.

[Symptoms]

The same error occurs in Teams, Edge Copilot, and the Copilot Studio test console.

Are other agents in the same environment working? → All agents in the environment are returning the same error message.
- Even a newly created agent produces the same error.

Agents in other environments are working normally.

[Environment Details]
The error is occurring only in a specific environment that is configured with the Early Release cycle. The issue started on June 11.

スクリーンショット 2026-06...

Your file is currently under scan for potential threats. Please wait while we review it for any viruses or malicious content.

Categories:

General topics

Good news: this isn't something you broke, and there's nothing to fix on your side. Your own findings point straight at a Microsoft-side issue. Every agent in the environment fails, including brand-new ones, and only this environment is affected while your others work fine. That can't be a config problem, it's the platform.

The Early Release cycle is the key detail. Those environments get platform updates ahead of the standard ones, so when Microsoft ships a bad change, that ring breaks first. That's why only this environment is down while your standard-cycle ones are fine.

And you're not alone. Others are reporting the same SystemError starting June 11, and Microsoft Support has acknowledged it's affecting multiple customers, so it's a known service-side incident.

What I'd do:

Don't rebuild or reconfigure your agents. Since new ones fail too, it won't help.
Check Service health and the Message center in the Power Platform or Microsoft 365 admin center for an advisory.
Open or reference a support ticket with your Conversation ID and UTC time, noting that all agents fail only in your Early Release environment and others are seeing it since June 11. That routes it as the platform incident, not a config issue.
If it's urgent, run the agent from one of your working standard-cycle environments as a stopgap until the Early Release ring is patched.

The fix will come from Microsoft rolling back or patching the rollout, so keep an eye on the service health dashboard. Hope that clears it up.