Loss of meaningful human control

https://taxonomy.eticas.ai/risk/loss-of-human-control

Maturity: emerging

AI systems operating at speeds or complexity levels that make meaningful human oversight impractical, including inability to understand, interrupt, or reverse autonomous decisions.

This subcategory is emerging. It has not yet been validated through established assessment methods.

System type: ADM and LLM systems
Lifecycle stages: Post Processing

Mappings to external frameworks

Standards & frameworks

Framework Reference
EU AI Act (Regulation 2024/1689) Article 14 — Human oversight (effective oversight)
AIUC-1 — AI Underwriting Company Standard Flag high-risk outputs for human review

Taxonomies & vocabularies

Framework Reference
MIT AI Risk Repository Loss of human agency and autonomy