Sustainability & AI
Responsible AI
Principles and practices ensuring AI is developed and used in ethical, accountable and socially beneficial ways.
Definition
Principles and practices intended to ensure AI is developed and used in ways that are ethical, accountable and beneficial, encompassing fairness, transparency, privacy, safety and — increasingly — environmental impact.
Quick reference
At a glance
- Subject
- Sustainability & AI
- Editorial status
- Editorial draft
- Definition status
- Multiple accepted definitions
- Last updated
- 21 August 2026
References
This source provides part of the technical or institutional basis for the definition.
This source supports the explanation of how the term is applied, measured or governed in practice.
This source supports the wider sustainability significance and context described in the entry.
Overview
What it means
Responsible AI matured from corporate principles documents into operational programmes: review boards, impact assessments, model documentation and red-teaming. Frameworks such as the OECD principles, UNESCO's Recommendation and ISO/IEC 42001 give it intergovernmental and standards form; environmental sustainability is an emerging pillar.
How it is used
Organisations run responsible-AI programmes as management systems — increasingly certified to ISO/IEC 42001 — covering use-case review, risk assessment and monitoring of deployed systems, with disclosure expectations rising under the EU AI Act.
Why it matters
The phrase is everywhere and means everything from legal compliance to marketing posture. Its value depends on operationalisation: without assessment processes, documentation and accountability owners, 'responsible AI' is itself a claim requiring substantiation.
Review
Help keep this definition useful and accurate.
Submitted reviews stay private until an editor decides whether to accept and attribute them.
Endorse this definition
Confirm what works
Endorse the current wording when it is accurate and useful. Suggested changes use the separate editorial form below.