A Human Rights-Based Approach to Impact Assessment: Guide 3 of the Responsible AI Practitioner Guides for Taking a Human Rights-Based Approach to Generative AI
Resource Key: AKVM4DJD
Document Type: Report
Creator:
Creators Name:
Place:
Institution: Business for Social Responsibility (BSR)
Date: February 2025
Language: en
This paper provides guidance on how Responsible AI practitioners can integrate human rights principles and human rights assessment (HRA) methodology into other forms of impact assessments of generative AI (genAI) and how they can effectively scale impact assessment processes across their organizations. It includes the following sections: 1 Human Rights Assessment Methodology: Describes what human rights assessments are, their core methodological components, how they are conducted in practice, and their benefits and limitations for assessing risks to people and society associated with AI. 2 Integrating Human Rights into Other AI Impact Assessments: Describes how human rights principles and assessment methodology can be integrated into algorithmic audits/risk/impact assessments, model evaluations, fairness testing, data quality reviews, and red teaming. 3 Scaling Impact Assessment Processes: Describes the role of standalone vs. inte grated impact assessments, and how AI impact assessment processes can be effec tively scaled across organizations. 4 Integrating Human Rights Assessment Across the Value Chain: Provides guidance for how each entity of the value chain can integrate human rights into its impact assessment processes. 5 Key Resources