
[100% Off] Ai Supply Chain Security: Models, Data, Tools And Vendors
Secure third-party models, datasets, MCP servers and AI vendors, with hands-on labs on free tools you can run locally
What you’ll learn
- Explain why model files can run code when they load, and choose safer formats and loaders
- Scan, pin, sign and verify models, and explain what a signature does not prove
- Protect training data with quarantine, hashes and provenance records
- Harden AI libraries and pipelines after real compromises such as LiteLLM and PyTorch nightly
- Vet MCP servers, skills and plugins, and detect a tool that changes after approval
- Build and validate an AI bill of materials in CycloneDX
- Assess AI vendors and check model licences
- Run one intake gate that combines all of it
Requirements
- Basic Python and command-line comfort
- A computer that can run Python 3.11 or 3.12
- All labs use free tools, with no paid account, no GPU and no Docker
- No machine learning maths
Description
This course contains the use of artificial intelligence.
Your AI system is mostly made of things your team did not write: a downloaded model, a third-party dataset, open-source libraries, MCP servers, agent skills and a vendor API. Each one can arrive broken or hostile, and the tools you use for ordinary code see only part of the problem.
In this course we secure those parts with a fictional insurer, Ollerwick Mutual, as it takes a claims assistant from a quick experiment to an approved system. We start from documented incidents and from current guidance: the March 2026 joint guidance on AI and machine learning supply chain risks, the G7 minimum elements for an AI software bill of materials, OWASP, MITRE ATLAS, OpenSSF Model Signing and SLSA.
You will build harmless test files yourself, then run real open-source tools against them on your own machine. We look just as closely at what those tools miss. In our own tests, two model scanners returned a pass on a test file that a third flagged, and a tool scanner passed a hidden instruction once we reworded it. That is why we teach layers of checks, and what each check proves.
What you will be able to do:- Explain why model files can run code when they load, and choose safer formats and loaders- Scan, pin, sign and verify models, and explain what a signature does not prove- Protect training data with quarantine, hashes and provenance records- Harden AI libraries and pipelines after real compromises such as LiteLLM and PyTorch nightly- Vet MCP servers, skills and plugins, and detect a tool that changes after approval- Build and validate an AI bill of materials in CycloneDX- Assess AI vendors and check model licences- Run one intake gate that combines all of it
The course includes section quizzes, assignments, a practice test and downloads: an AI bill of materials template, a vendor questionnaire and an intake checklist.
This is not a logistics course. We cover digital parts of AI systems. We do not cover freight, warehouses or ISO 28000 physical supply chain security, and we do not teach AI for logistics planning.
Author(s): Dr. Amar Massoud








