OSTP Releases Artificial Intelligence Bill of Rights
The White House Office of Science and Technology Policy (OSTP) recently released a Blueprint for an Artificial Intelligence Bill of Rights aimed at protecting diverse communities through identifying concerns, risks, and potential solutions to combat discrimination and biases with artificial intelligence systems.
To combat the rise of inequity and biases with Artificial Intelligence (AI) technology, OSTP’s Bill of Rights blueprints identifies five practices:
- Safe and Effective Systems: Designers, developers, and deployers of automated systems should ensure systems have pre-deployment testing that identifies risks and determines the safety of the system.
- Algorithmic Discrimination Protections: Designers, developers, and deployers of automated systems should protect communities from algorithmic discrimination using proactive equity assessment, representative data, and protection against proxies for demographic features.
- Data Privacy: Designers, developers, and deployers of automated systems should require permission in a brief and understandable way to use data and safeguard the data when applicable to the best of their ability.
- Notice and Explanation: Designers, developers, and deployers of automated systems should provide clear and easy-to-understand explanations on what systems are being used and what they are being used for.
- Human Alternatives, Consideration, and Fallback: When possible, there should be alternatives to automated systems to not only replace the system but to also solve any problems encountered by the user.
The blueprint is intended as a guide to protect the public’s privacy and use artificial intelligence responsibly. In addition, the White House released an online handbook, from Principles to Practice, aimed at assisting with the incorporation of these principles into policy and practice.