← Back to ATLAS MatrixTactic 10 of 15

Defense Evasion

The adversary is trying to avoid being detected in your AI system.

7 Techniques
MITRE ATLAS Framework

Attack Techniques

Explore the specific techniques adversaries use during the defense evasion phase of AI attacks.

1
AML.T0015

Impair Defenses

Adversaries may modify system configurations to impair defensive capabilities in AI systems.

Examples:

  • •Disabling AI security monitoring
  • •Manipulating model validation
  • •Bypassing detection mechanisms

Mitigations:

  • ✓Immutable security controls
  • ✓Regular security audits
  • ✓Defense in depth

References:

MITRE ATLAS AML.T0015
2
AML.T0067

Impair Defenses

Adversaries may modify system configurations to impair defensive capabilities in AI systems.

Examples:

  • •Disabling AI security monitoring
  • •Manipulating model validation
  • •Bypassing detection mechanisms

Mitigations:

  • ✓Immutable security controls
  • ✓Regular security audits
  • ✓Defense in depth

References:

MITRE ATLAS AML.T0067
3
AML.T0068

Impair Defenses

Adversaries may modify system configurations to impair defensive capabilities in AI systems.

Examples:

  • •Disabling AI security monitoring
  • •Manipulating model validation
  • •Bypassing detection mechanisms

Mitigations:

  • ✓Immutable security controls
  • ✓Regular security audits
  • ✓Defense in depth

References:

MITRE ATLAS AML.T0068
4
AML.T0071

Impair Defenses

Adversaries may modify system configurations to impair defensive capabilities in AI systems.

Examples:

  • •Disabling AI security monitoring
  • •Manipulating model validation
  • •Bypassing detection mechanisms

Mitigations:

  • ✓Immutable security controls
  • ✓Regular security audits
  • ✓Defense in depth

References:

MITRE ATLAS AML.T0071
5
AML.T0073

Impair Defenses

Adversaries may modify system configurations to impair defensive capabilities in AI systems.

Examples:

  • •Disabling AI security monitoring
  • •Manipulating model validation
  • •Bypassing detection mechanisms

Mitigations:

  • ✓Immutable security controls
  • ✓Regular security audits
  • ✓Defense in depth

References:

MITRE ATLAS AML.T0073
6
AML.T0074

Impair Defenses

Adversaries may modify system configurations to impair defensive capabilities in AI systems.

Examples:

  • •Disabling AI security monitoring
  • •Manipulating model validation
  • •Bypassing detection mechanisms

Mitigations:

  • ✓Immutable security controls
  • ✓Regular security audits
  • ✓Defense in depth

References:

MITRE ATLAS AML.T0074
7
AML.T0076

Impair Defenses

Adversaries may modify system configurations to impair defensive capabilities in AI systems.

Examples:

  • •Disabling AI security monitoring
  • •Manipulating model validation
  • •Bypassing detection mechanisms

Mitigations:

  • ✓Immutable security controls
  • ✓Regular security audits
  • ✓Defense in depth

References:

MITRE ATLAS AML.T0076
← Privilege EscalationATLAS MatrixCredential Access →