User Execution
Description
An adversary may rely upon a user to load, execute, interpret, or otherwise use a malicious or unsafe artifact. In AI systems, user execution may include loading a model or causing an inference runtime to process executable or behavior-shaping components bundled with an AI artifact. The resulting effect may include host code execution or malicious AI behavior.
Users may inadvertently execute unsafe code introduced via AI Supply Chain Compromise.
Users may be subjected to social engineering to get them to execute malicious code by, for example, opening a malicious document file or link.
- ATLAS ID
- AML.T0011
- Maturity
- RealizedUsed by real adversaries against AI systems in the wild.
- Platforms
- Predictive AIGenerative AIAgentic AIEnterprise
Sub-techniques
Mitigations
- AML.M0011Restrict Library Loading
Restricting binaries from loading external libraries can limit their ability to execute malicious code.
- AML.M0014Verify AI Artifacts
Introduce proper checking of signatures to ensure that unsafe AI artifacts will not be executed in the system.
- AML.M0016Vulnerability Scanning
Vulnerability scanning can help identify malicious binaries and prevent user execution.
- AML.M0018User Training
Training users to be able to identify attempts at manipulation will make them less susceptible to performing techniques that cause the execution of malicious code.
- AML.M0023AI Bill of Materials
An AI BOM can help users identify untrustworthy binaries.