AI Gateway Adds Team-Level Provider Allowlist Support

AI Gateway now supports an Allowlist feature that restricts which AI providers a team can use. This ensures traffic is routed only to approved vendors that have passed security and legal review.

Key features include:

• Gateway-level enforcement: Even if a developer or coding agent modifies filters at the request level, routing to unapproved providers is blocked.
• Centralized control: Only team owners can modify the allowlist, making it easy to manage and audit.
• Default-deny policy: When the allowlist is enabled, new providers are disabled by default, preventing unintended vendor expansion.

This feature filters at the provider level rather than the model level, and works across all API formats supported by AI Gateway, including AI SDK, OpenAI Chat Completions API, and Anthropic Messages API. It also applies alongside existing security settings such as Zero Data Retention (ZDR) to build an even stronger security posture.