offensive-ai-compilation

Free

A curated list of useful resources that cover Offensive AI.

CybersecurityFreeFree tier

Type

Open Source

Visit Website GitHub (1.4k stars)

About offensive-ai-compilation

A comprehensive, curated compilation of resources covering Offensive AI, organized by attack types (extraction, inversion, poisoning, evasion), defensive actions, tools (e.g., ART, Cleverhans), and use cases including pentesting, malware, OSINT, phishing, and generative AI. The list includes links to papers, frameworks, and surveys, serving as a hub for researchers and practitioners.

Key Features

Curated list of offensive AI resources

Coverage of four attack types: extraction, inversion, poisoning, evasion

Defensive actions and countermeasures for each attack type

Tools such as ART (Adversarial Robustness Toolbox) and Cleverhans

Use cases: pentesting, malware, OSINT, phishing, generative AI (audio, image, video, text)

Includes surveys and links to academic papers

Open source and freely accessible

Pros & Cons

Pros

Comprehensive collection of resources across multiple offensive AI domains
Well-organized into categories with defensive actions included
Open source and free to use
Covers both theoretical concepts and practical tools
Regularly maintained with links to recent papers

Cons

Not an interactive tool but a static list of resources
No built-in functionality beyond navigation
May require manual effort to keep links up to date
Some external links may become outdated

Best For

Adversarial machine learning researchPenetration testing with AIMalware analysis and developmentOSINT gathering and reconnaissancePhishing attack simulation and detectionGenerative AI exploitation (audio, image, video, text)Model extraction and stealingMembership inference and property inference

Alternatives to offensive-ai-compilation

giskard-oss

🐢 Open-Source Evaluation & Testing library for LLM Agents

ai-captcha-bypass

AI Captcha Bypass

backdoor-learning-resources

A list of backdoor learning resources

claude-bug-bounty

AI-powered bug bounty hunting from your terminal - recon, 20 vuln classes, autonomous hunting, and report generation. All inside Claude Code.

agentic-radar

A security scanner for your LLM agentic workflows

FAQ

What is Offensive AI?

Offensive AI involves exploiting the vulnerabilities of AI models, including attacks like extraction, inversion, poisoning, and evasion.

What categories are covered in this compilation?

The compilation covers abuse, adversarial machine learning, four types of attacks (extraction, inversion, poisoning, evasion), defensive actions, tools (such as ART and Cleverhans), and use cases including pentesting, malware, OSINT, phishing, and generative AI.

Is this tool actively maintained?

The compilation appears to be maintained on GitHub (based on the Jekyll theme reference), but specific update frequency is not stated.