We take security seriously at AI Runway. We appreciate your efforts to responsibly disclose your findings and will make every effort to acknowledge your contributions.

Please do NOT report security vulnerabilities through public GitHub issues.

Instead, please report them via GitHub Security Advisories.

If you prefer email, you can send your report to the repository maintainers. Please include the word "SECURITY" in the subject line.

You should receive a response within 72 hours. If for some reason you do not, please follow up to ensure we received your original message.

Please include as much of the following information as possible to help us better understand the nature and scope of the issue:

• Type of issue (e.g., unauthorized access, privilege escalation, code injection, data exposure, etc.)
• Full paths of source file(s) related to the manifestation of the issue
• Location of the affected source code (tag/branch/commit or direct URL)
• Any special configuration required to reproduce the issue
• Step-by-step instructions to reproduce the issue
• Proof-of-concept or exploit code (if possible)
• Impact of the issue, including how an attacker might exploit it

• Acknowledgment: We will acknowledge receipt of your vulnerability report within 72 hours.
• Triage: We will confirm the problem and determine the affected versions within 7 days.
• Fix & Disclosure: We will work on a fix and coordinate disclosure timing with you. We aim to release patches within 30 days of triage confirmation.
• Credit: We will credit reporters in the advisory (unless you prefer to remain anonymous).

We follow a coordinated disclosure model:

1. The reporter submits the vulnerability privately.
2. We confirm and triage the vulnerability.
3. We develop and test a fix.
4. We release the fix and publish a security advisory.
5. We publicly disclose the vulnerability details after the fix is available.

We ask that you:

• Allow us a reasonable amount of time to address the issue before public disclosure.
• Make a good faith effort to avoid privacy violations, destruction of data, and interruption or degradation of our services.
• Do not access or modify other users' data.

Security advisories will be published via GitHub Security Advisories.

The following are considered in-scope for security reports:

• Controller vulnerabilities: Issues in the Kubernetes controller that could allow unintended resource creation, modification, deletion, or privilege changes
• API and dashboard vulnerabilities: Authentication, authorization, input validation, or injection issues in the backend API, web UI, or Headlamp plugin
• Custom resource validation bypasses: Ways to bypass validation for ModelDeployment, InferenceProviderConfig, or related resources that could cause unsafe deployments
• Provider integration issues: Vulnerabilities in AI Runway provider integrations that could expose credentials, model data, or cluster resources
• Denial of service: Resource exhaustion attacks that bypass configured limits or disrupt deployments managed by AI Runway
• Supply chain attacks: Vulnerabilities in dependencies, build artifacts, or release assets that affect AI Runway

The following are generally out of scope:

• Issues in third-party inference providers, models, or Kubernetes distributions that are not caused by AI Runway (report to the respective maintainers)
• Social engineering attacks
• Denial of service through normal usage within configured resource limits

We recommend always running the latest version of AI Runway to benefit from the most recent security patches.

• Pin dependencies: Use lockfiles to ensure reproducible builds.
• Review generated manifests: Inspect generated deployment manifests before applying them to shared or production clusters.
• Use least-privilege credentials: Grant AI Runway only the cluster permissions required for the providers and namespaces you use.
• Keep updated: Regularly update AI Runway to receive security fixes.
• Review provider access: Only configure inference providers and model sources that you trust.

This security policy is based on the CNCF TAG Security project resources templates.