Research examining responsible AI, trustworthy AI, and ethical AI reflects a growing awareness of the need to address ethical considerations and societal impact in the development and deployment of artificial intelligence systems.  Equitable AI, responsible, ethical, and trustworthy AI, while often considered synonymous and used interchangeably, possess key differences and inform the goals of this guide. 

‍

Responsible AI

Responsible AI addresses the overall approach to AI development and deployment, including societal impacts and considerations such as fairness, transparency, accountability, privacy, and safety.

‍

Trustworthy AI

Trustworthy AI is more specific in focus concerned with designing and implementing AI systems that are reliable, secure, and transparent, fostering trust among users and stakeholders. 

‍

Ethical AI

Ethical AI narrows the focus even more to fairness, accountability, transparency, privacy, and human rights.

‍

Equitable AI

Equitable AI, the most specific, is the design and development of AI systems that are fair and unbiased, ensuring that all individuals have equal access to its benefits and are not disproportionately harmed by its potential risks.

‍

‍

Equitable AI shares the scope of the others and focuses specifically on fairness, avoiding bias, and mitigating the devastating consequences of inequitable AI. This guide focuses on equitable AI-promoting opportunities and outcomes for all user groups.