AI Supply Chain Compromise
Description
Adversaries may gain initial access to a system by compromising the unique portions of the AI supply chain.
This could include Hardware, Data and its annotations, parts of the AI Software stack, or the Model itself.
In some instances the attacker will need secondary access to fully carry out an attack using compromised components of the supply chain.
- ATLAS ID
- AML.T0010
- Maturity
- RealizedUsed by real adversaries against AI systems in the wild.
- Platforms
- Predictive AIGenerative AIAgentic AI
Sub-techniques
Mitigations
- AML.M0014Verify AI Artifacts
Introduce proper checking of signatures to ensure that unsafe AI artifacts will not be introduced to the system.
- AML.M0020Generative AI Guardrails
Guardrails can detect harmful code in model outputs.
- AML.M0023AI Bill of Materials
An AI BOM can help users identify untrustworthy components of their AI supply chain.
- AML.M0035AI Red Team
Exercise the introduction of controlled untrusted software, data, models, and agent tools through representative acquisition and deployment paths. Remediate weaknesses in provenance, validation, approvals, isolation, and rollback.