Content Safety in Foundry Control Plane
An Azure service that enables users to identify content that is potentially offensive, risky, or otherwise undesirable. Previously known as Azure Content Moderator.
Intermittent "violence: high" content filter block on a benign appointment-booking chat – Azure OpenAI Title: Intermittent "violence: high" content filter block on a benign appointment-booking chat – Azure OpenAI
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(128, 44%, 22%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ENN%3C/text%3E%3C/svg%3E)
Nishyanth Nandagopal
0
Reputation points
Hi all,
We run customer-facing chatbots on Azure OpenAI, orchestrated with Semantic Kernel (Python). Some of them intermittently get a prompt-side content filter block in the violence category (severity "high"), even though the conversation is an ordinary appointment booking. The same flow usually works when retried later.
ERROR
HTTP 400, code "content_filter", param "prompt",
innererror "ResponsibleAIPolicyViolation".
content_filter_result: violence filtered (high); hate, sexual, self_harm
all "safe"; jailbreak not detected.
CONVERSATION WHEN IT FAILED
User: hi -> greeting
User: asks to book an appointment -> bot asks for name and email
User: provides name and email -> bot asks for preferred time
User: "tomorrow 10 am" -> request blocked
Thanks!
Content Safety in Foundry Control Plane
1 answer
Sort by: Most helpful
-
Deleted
This answer has been deleted due to a violation of our Code of Conduct. The answer was manually reported or identified through automated detection before action was taken. Please refer to our Code of Conduct for more information.
Comments have been turned off. Learn more