huntr.com

A comprehensive bug bounty platform dedicated to AI and machine learning open-source applications, libraries, and model file formats.

Visit Site

AI Developer Tools Open Source AI Models AI Models

About huntr.com

huntr is the pioneering bug bounty platform tailored for AI and ML, offering a centralized portal for security researchers to report vulnerabilities. It ensures the security and reliability of open-source AI/ML applications, libraries, and model file formats through structured vulnerability disclosure and bounty programs.

How to Use

Security researchers submit vulnerabilities via a secure form. huntr verifies reports and contacts maintainers, who have 31 days to respond. Valid reports result in bounties, with public disclosures happening after 90 days unless an extension is requested. Model file format reports are kept confidential.

Features

Structured vulnerability submission system

Specialized focus on AI and ML security

Secure and transparent disclosure process

Bug bounty program administration

CVE assignment for open-source vulnerabilities

Use Cases

Reward security researchers for discovering vulnerabilities

Secure AI/ML model file formats effectively

Identify and address vulnerabilities in open-source AI projects

Enhance the security of AI and ML applications

Best For

MLSecOps engineersSecurity researchersOpen-source project maintainersAI/ML developers

Pros

Provides CVE identifiers for open-source vulnerabilities

Incentivizes security research with bug bounties

Facilitates collaboration between researchers and maintainers

Offers a structured vulnerability disclosure platform

Specialized focus on AI/ML security challenges

Cons

Model file format reports are not publicly disclosed

Responders have 31 days, which may delay issue resolution

Current lack of support for researchers submitting patches or claiming fix bounties

Frequently Asked Questions

Find answers to common questions about huntr.com

How does the vulnerability disclosure process work?

Researchers submit vulnerabilities via a secure form. huntr verifies reports and contacts maintainers, who have 31 days to respond. Valid reports result in bounties, with public disclosure after 90 days unless extended. Model file format reports remain confidential.

What if a maintainer does not respond to a vulnerability report?

If no response is received within 31 days, huntr will resolve high and critical reports within 14 days.

When do open-source vulnerability reports become public?

All open-source vulnerability reports are publicly disclosed after 90 days, unless maintainers request an extension. Informational or invalid reports are disclosed immediately.

Can researchers submit patches or claim fixes?

Currently, the platform does not support patch submissions or fix bounty claims from researchers.

Is vulnerability reporting limited to specific AI/ML models?

No, the platform supports reporting vulnerabilities across various AI/ML applications, libraries, and model file formats.