Modify AI Model Architecture
Sub-technique of AML.T0018 Manipulate AI Model
Description
Adversaries may directly modify an AI model's architecture to re-define it's behavior. This can include adding or removing layers as well as adding pre or post-processing operations.
The effects could include removing the ability to predict certain classes, adding erroneous operations to increase computation costs, or degrading performance. Additionally, a separate adversary-defined network could be injected into the computation graph, which can change the behavior based on the inputs, effectively creating a backdoor.
- ATLAS ID
- AML.T0018.001
- Maturity
- DemonstratedShown to work in research or red-team exercises.
- Platforms
- Predictive AIGenerative AIAgentic AI
Mitigations
- AML.M0005Control Access to AI Models and Data at Rest
Access controls can prevent tampering with ML artifacts and prevent unauthorized copying.
- AML.M0008Validate AI Model
Ensure that acquired models do not respond to potential backdoor triggers or adversarial influence.
- AML.M0013Code Signing
Code signing provides a guarantee that the model has not been manipulated after signing took place.
- AML.M0035AI Red Team
Attempt controlled unauthorized changes to model architecture or executable model components. Verify review, integrity checking, signing, deployment approval, and restoration controls.