The US Government’s objectives in governing AI are best articulated in the White House’s ‘Blueprint for an AI Bill of Rights’ (2022), which aims to serve as a guide for ‘a society that protects all people from these threats—and uses technologies in ways that reinforce our highest values’:
• systems should undergo pre-deployment testing, risk identification and mitigation, and ongoing monitoring.
• protect individuals and communities from algorithmic discrimination and ensure the use and design of systems in an equitable way.
• stricter data practices compel AI technologies to seek permission before collecting, using, accessing, transferring and deleting personal data.
introduce documentation and explanation around when and how users interact with an automated system.
• offer an opt-out and the option of a human alternative.
On top of those measures, the US Government is invested in maintaining and protecting its edge in frontier-technology AI systems (those technologies that can perform in a way that matches or exceeds capabilities present in today’s most advanced models).
Many of those objectives were subsequently reinforced through Executive Order 14110