High‑risk application
AI uses with significant impacts on life, rights, or critical infrastructure.
How the laws define this
Regulators uniformly view high-risk AI applications as systems whose deployment, failure, or error creates significant risks to fundamental rights, personal safety, health, or critical infrastructure, thereby requiring mandatory enhanced governance and risk mitigations. The primary divergence lies in which specific sectors trigger this classification and the precise administrative burdens imposed. The National Artificial Intelligence Policy (National AI Policy) 2025 explicitly enumerates precise contexts like health diagnostics, social services, and law enforcement, while pairing oversight with mandatory registration. Similarly, the Buku Putih Peta Jalan Kecerdasan Artifisial Nasional (White Paper / Roadmap for National AI) extends scope to economic livelihood and potential registration. Conversely, A Blueprint for Greece's AI Transformation uniquely conditions its strict oversight, documentation, and conformity assessments on EU-aligned risk-based regulation. Additionally, the National Artificial Intelligence Policy explicitly incorporates national security threats and mandates audits and testing, whereas the National Guidelines on AI Governance & Ethics (AIGE) relies on broader, illustrative sectoral categories.
Synthesised from the 5 statutory definitions below. Each one is quoted in full, with its source.
Definitions (5)
Applications or uses of AI that significantly affect life, health, fundamental rights, critical infrastructure, financial stability, or national security, and which therefore trigger enhanced regulatory measures such as mandatory impact assessments, testing, certification, audits, and sectoral safeguards.
AI systems whose failure, error, or misuse could materially affect safety, fundamental rights, economic outcomes, or other serious interests (e.g., in healthcare, finance, critical infrastructure), and which therefore warrant heightened assessment and controls.
AI applications deployed in contexts that materially affect rights, safety, or critical outcomes — expressly including health diagnostics, critical infrastructure, social services, law enforcement, and financial decision‑making — and therefore subject to heightened compliance, registration, and oversight requirements.
Applications or uses of AI that have significant potential to affect fundamental rights, public health, safety, or economic livelihood; these uses are subject to additional risk assessment, mitigation measures, and potential registration or notification requirements.
AI systems or uses that, by virtue of their function, sector, or potential impact, create significant risks to fundamental rights, safety, or essential public functions; these are targeted for stricter oversight, documentation and conformity assessment in line with EU-aligned risk-based regulation.
Related Terms
High-risk applications
Uses that affect fundamental rights, health, safety, or livelihoods....
high-risk application
AI use in contexts that materially affect rights or safety....
High‑impact uses
Uses that materially affect rights, health, safety, or livelihoods....
high-risk use cases
Public‑interest deployments with significant impact on rights or services....
High-Risk (AI)
AI uses that pose substantial potential harm....