Under active development Content is continuously updated and improved

Art. 15Accuracy, Robustness and Cybersecurity

>Control Description

High-risk AI systems shall be designed and developed to achieve appropriate levels of accuracy, robustness and cybersecurity, and perform consistently throughout their lifecycle. Accuracy levels shall be declared in instructions for use and systems shall be resilient against errors, faults, inconsistencies, and malicious third-party attempts.

Risk Tier

High-Risk

Article

Art. 15

Deadline

August 2, 2026

Penalty

€15M or 3% of global turnover

>Applies To

providers

>Key Requirements

  • Appropriate levels of accuracy for intended purpose
  • Accuracy levels declared in instructions for use
  • Resilient to errors, faults, or inconsistencies
  • Robust against unauthorized third-party attempts
  • Protection against data poisoning and adversarial attacks
  • Appropriate cybersecurity measures
  • Fail-safe mechanisms for continuous learning systems

>Cross-Framework Mappings

Ask AI

Configure your API key to use AI features.