Skip to content
cisonews.ai
US FederalPublished

White House OSTP releases the Blueprint for an AI Bill of Rights

On 4 October 2022 the White House OSTP released the Blueprint for an AI Bill of Rights, five non-binding protections for people affected by automated systems.

Date
2022-10-04
Instrument
Blueprint for an AI Bill of Rights
Published
Sources
1

Compiled September 27, 2026 from public records. The events described took place on October 4, 2022.

By the cisonews.ai Desk · Published

The Blueprint for an AI Bill of Rights is a White House framework setting out five protections that people in the United States should have when automated systems affect them, and the Office of Science and Technology Policy (OSTP) released it on 4 October 2022. It is guidance for the design, development and deployment of AI and other automated systems rather than a statute, and the page announcing it now sits on the archived Biden White House site.

What it is

OSTP said it developed the framework over a year-long process that drew on panel discussions, public listening sessions, meetings and a formal request for information. The blueprint comes with a technical companion that gives examples and concrete steps for communities, industry and governments. OSTP suggested that project managers building AI products use it as a checklist and that policymakers use it to develop sector guidance on automated systems.

What it requires

As a blueprint, it creates expectations rather than legal obligations. Its five principles are safe and effective systems, protection from algorithmic discrimination, data privacy with protection from abusive data practices, notice and explanation when an automated system is in use, and access to human alternatives, consideration and fallback, including the ability to opt out where appropriate. The data privacy principle calls for built-in protections and for people to have agency over how data about them is used.

What security leaders should do

Read it as an early statement of what US policymakers expected from automated systems, not as a current compliance duty. Several principles overlap with security work: data minimisation and access controls support the privacy principle, testing and monitoring support safe and effective systems, and decision logs support notice and explanation. Later instruments, including state AI laws tracked on this site, pick up similar themes, so controls built around these five headings tend to carry over.

Sources (1)

  1. [1]
    White House Office of Science and Technology Policy (archived)
    bidenwhitehouse.archives.govOctober 4, 2022tier A · Primary

Spotted an error? Request a correction. Every fact on this page is tied to a listed source.