AI Azure Content Safety

Release date 2026/07/02

Bugfix

• Fixed a crash (HTTP 500) that occurred when azure_use_managed_identity was enabled.

• Fixed a bug where the plugin returned HTTP 500 when the LLM response content was empty. Empty content now skips the content-safety check and passes through.

• Fixed an issue where the response guard was sending the entire JSON response body to Azure Content Safety instead of just the assistant message content, which caused false positives when response metadata fields matched blocklist entries.

Release date 2026/05/25

Bugfix

• Fixed a crash (HTTP 500) when azure_use_managed_identity is enabled.

Release date 2026/04/07

Feature

• the ssl_verify option to verify the certificate presented by the Azure Content Safety service when using HTTPS is now enabled by default.

Release date 2026/06/30

Bugfix

• Fixed a crash (HTTP 500) when azure_use_managed_identity is enabled.

Release date 2025/12/18

Feature

• Enabled at-rest keyring encryption for sensitive fields in AI Azure Content Safety plugin.

• Added support for the global tls_certificate_verify option. When enabled globally via kong.conf, the plugin’s ssl_verify config field cannot be set to false. This ensures SSL/TLS certificate verification cannot be disabled when global security policy requires it.

Release date 2025/10/01

Feature

• Added response guard to filter AI responses for safety and compliance.

Bugfix

• Fixed the plugin to work with multi-modal content.

Release date 2025/07/16

Bugfix

• Fixed an issue where the llm license migration could fail if the license counter contained more than one week of data.

Release date 2025/07/03

Bugfix

• AI Azure Content Safety: Fixed an issue where Blocklist breaches are not correctly parsed.

Known Issues

• If any AI Gateway plugin has been enabled in a self-managed Kong Gateway deployment for more than a week, upgrades from 3.10 versions to 3.11.0.0 will fail due to a license migration issue. This does not affect Konnect deployments.

  A fix will be provided in 3.11.0.1.

  See breaking changes in 3.11 for a temporary workaround.

Release date 2025/08/07

Bugfix

• AI Azure Content Safety: Fixed an issue where Blocklist breaches are not correctly parsed.

Release date 2024/05/28

Feature

• Adds a new plugin that allows the Kong administrator to enforce that all AI-Proxy requests must be introspected with the Azure Content Safety service.

  The plugin enables configurable thresholds for the different moderation categories, and reports audit results into the Kong log serializer for reporting purposes.