AI Safety Testing Program Brief

Introduction

Welcome to the Genbounty AI Safety Program. Through this program, we aim to ensure our AI systems align with ethical guidelines, safety standards, and fairness principles. We invite security researchers, AI experts, and human testers to help us identify and mitigate potential risks in our AI models and systems.

About Genbounty

Generate an AI safety program for your project, empower users to privately disclose AI safety issues to your team. Comply with EU AI Act regulations. Crowdsource AI testers and bug bounty hunters to test your app.

Scope

In-Scope Systems & Models

AI application testing and evaluation
Safety and alignment assessment
Ethical considerations and bias detection
User interaction analysis
AI systems at genbounty.com

Out-of-Scope Systems

Standard software vulnerabilities unrelated to AI
Network security testing
Physical security testing
Social engineering attacks

What to Report

Bias & Fairness Issues: AI models displaying or reinforcing harmful biases
Security Risks: Prompt injection, model manipulation, or data poisoning
Hallucination & Misinformation: AI generating misleading or harmful outputs
Privacy Violations: Unintended exposure of user data or model training data
Autonomy & Control Issues: AI taking unintended actions beyond its designed capabilities

Submission Guidelines

Provide clear and detailed feedback about the identified issue
Include reproducible steps or a proof-of-concept
Explain the potential impact and suggest possible mitigations
Submit through the official feedback form

Responsible Disclosure

Do not publicly disclose the vulnerability before we resolve the issue
Only test systems within the defined scope
Do not exploit vulnerabilities beyond what is necessary to demonstrate the issue

Legal Considerations

Genbounty reserves the right to determine eligibility and rewards. Participants must comply with applicable laws and ethical guidelines.

Contact

For any inquiries or clarifications, please reach out to programs@genbounty.com.

Thank you for helping us create safer and more aligned AI systems!