From Model To Trust: Building Upon Tamper-proof ML Metadata Records
summary
Demonstrates a working system that cryptographically signs bundles of ML artifacts (models, datasets, model cards) using OpenSSF Model Signing, produces short‑lived certificates via an OIDC CA (demo shows certificates valid for ~10 minutes / ~10 seconds) and records signatures in a transparency log so verification fails on tampering. Uses PyTorch for training, a pip-installable model-signing CLI, and integrations with Kaggle and NVIDIA model hubs; demos show signing, upload, and local verification succeeding.
AI-written summary · about
chapters (14)
- 0:00Tamper-proof ML metadata
- 1:07AI supply chain security
- 3:38Model tampering risk demo
- 5:22Traditional checks fail for ML
- 7:10Model signing and attestations
- 9:38Model signing in practice
- 11:03Model signing workflow demo
- 14:00Transparency logs prevent impersonation
- 16:06Kaggle automated verification
- 18:56Sixstore vs traditional signing
- 20:20Nvidia’s model signing adoption
- 21:46Future of ML workflow signing
- 22:48Signing ML pipelines
- 23:51Contributing to the project
related talks
26mTaming ML’s Wild West: Practical Model Signing with Sigstore on Kaggle
27mTaming the Wild West of ML: Practical Model Signing with Sigstore on Kaggle
23mEnd-to-End Secure ML Development
21mML Model Signing: Cryptographically Paving the Way to Provenance in Machine Learning Models
31mBuilding Trust in ML: Mapping the Model Lifecycle for ML Integrity and Transparency
19mModel transparency for AI/ML security
21mEnd-to-end secure ML Development
23mTrust, Track, and Verify: Securing AI Pipelines End-to-End