AI 용어집
인공지능 완전 사전
Ethical Impact Assessment (EIA)
Systematic methodology for identifying, evaluating, and mitigating the potential ethical consequences of an AI system before and during its deployment.
Impact Assessment
Structured process of systematically analyzing the potential and actual effects of an AI technology on individuals and society.
Ethical Risk Assessment
Systematic identification of potential ethical threats and assessment of their likelihood of occurrence and severity.
Algorithmic Fairness Metrics
Quantitative indicators that measure the level of fairness and impartiality of an AI system across different demographic groups.
Differential Impact Analysis
Assessment of the varied effects of an AI system on different population subgroups to identify disparities.
Bias Mapping
Systematic process of identifying and documenting all potential sources of bias in the AI lifecycle.
Context Sensitivity Assessment
Analysis of the suitability of an AI system to the cultural, social, and contextual specificities of its application domain.
Algorithmic Social Impact
Measurement of the direct and indirect consequences of algorithmic decisions on social structure and human relationships.
Transparency Assessment
Analysis of the level of explainability and understandability of an AI system's decision-making processes for its users.
Fundamental Rights Impact Analysis
Specific evaluation of the effects of an AI on essential human rights such as non-discrimination, privacy, and freedom of expression.
Algorithmic Accountability Assessment
Method to determine clear responsibility chains in complex and automated AI systems.
Algorithmic Harm Indicators
Specific metrics designed to detect and quantify potential harm caused by automated decisions.
Cumulative Impact Assessment
Analysis of the combined and synergistic effects of multiple AI systems operating simultaneously on the same population.
Ethics by Design Framework
Methodological approach integrating ethical considerations from initial design and throughout AI development.
Privacy Impact Analysis
Specific evaluation of the risks and consequences of an AI system on personal data protection and user privacy.
Impact on Diversity and Inclusion
Measurement of the effects of an AI system on the representation and fair treatment of underrepresented or marginalized groups.
Intersectional Bias Assessment
Analysis of complex forms of discrimination emerging from the intersection of multiple identity characteristics in AI systems.