UTC

AML.T0043.001

Craft Adversarial Data: Black-Box Optimization

In Black-Box attacks, the adversary has black-box (i.e. AI Model Inference API Access via API access) access to the target model. With black-box attacks, the adversary may be using an API that the victim is monitoring. These attacks are generally less effective and require more inferences than White-Box Optimization attacks, but they require much less…

  • Tactic AI Attack Adaptation
  • Maturity: Demonstrated
  • Predictive AI
Full entry on MITRE ATLAS →
Mitigations
  • Predictive AI Output ObfuscationAML.M0002

    Reduce the fidelity and amount of information returned by predictive AI inference endpoints to make model discovery, extraction, replication, and black-box adversarial-example optimization more difficult.

    Technical - AI

  • Predictive AI Model HardeningAML.M0003

    Design and train predictive AI models to maintain intended performance when presented with adversarial examples. Adversarial examples may include digitally perturbed inputs or physical countermeasures intended to cause misclassification, missed detection, or another attacker-selected prediction.

    Technical - AI

  • Limit AI Service Query Volume and RateAML.M0004

    Limit the number and rate of requests that users can submit to an AI service. Apply limits by user, API key, tenant, device, or other authenticated identity. Use short-term rate, burst, and concurrency limits together with longer-term usage quotas.

    Technical - Cyber

  • Predictive AI EnsemblesAML.M0006

    Use an ensemble of diverse predictive AI models to reduce reliance on a single model or model family and improve robustness against adversarial examples.

    Technical - AI

  • Predictive AI Input RestorationAML.M0010

    Preprocess predictive AI inference inputs to remove, reduce, or disrupt adversarial perturbations before the inputs are evaluated by the model.

    Technical - AI

  • Predictive AI Adversarial Input DetectionAML.M0015

    Detect and block digital or physical adversarial examples submitted to predictive AI models. Adversarial examples are inputs modified or constructed to cause misclassification, missed detection, excessive computation, or another attacker-selected behavior.

    Technical - AI

  • Control Access to AI Models and Data in ProductionAML.M0019

    Require users to verify their identities before accessing a production model. Require authentication for API endpoints and monitor production model queries to ensure compliance with usage policies and to prevent model misuse.

    Policy

  • Adaptive AI Response DegradationAML.M0040

    Reduce the utility of AI model outputs for unauthorized model replication, knowledge distillation, or inference about training data by dynamically modifying responses when suspected extraction activity or privacy attacks are detected.

    Technical - AI

From MITRE ATLAS 2026.10, Copyright 2021-2026 MITRE, Apache License 2.0.