# Model Supply Chain Attacks

> Master the vulnerabilities of the AI supply chain. Learn how attackers compromise model repositories, exploit serialization flaws, and defend with Model SBOMs.

Canonical: https://codelivly.com/rooms/model-supply-chain-attacks
Last updated: 2026-10-12

| | |
|---|---|
| Level | Intermediate |
| Category | Cybersecurity |
| Lessons | 3 |
| Estimated time | 30 min |
| XP | 100 |

This module explores the critical vulnerabilities inherent in the AI software supply chain. You will learn how adversaries weaponize public model registries, exploit insecure serialization formats (like Python's pickle), and execute dependency confusion attacks to compromise inference environments. Through practical scenarios and defensive methodologies, you will master the implementation of Model Bills of Materials (Model SBOMs), cryptographic verification, and secure registry practices to protect your AI infrastructure.

## Lessons

1. What is the AI Model Supply Chain (30 min)
2. Risks of Downloading Models from Public Hubs (Hugging Face, etc.) (30 min)
3. Malicious Model File Formats (Pickle Exploits, Unsafe Serialization) (30 min)

## Related

- [Learning path: Data and Model Poisoning](https://codelivly.com/learning-paths/data-model-poisoning)
- [All rooms](https://codelivly.com/rooms)