The White House Publishes a “Blueprint” for an AI Bill of Rights to Protect US Citizens From Harm


The White House Office of Science and Technology Policy (OSTP) has announced a “blueprint for an AI Bill of Rights,” with five key principles for the design, use, and deployment of automated systems — in order to deliver on the promise of artificial intelligence technology without putting citizens at risk.

“Among the great challenges posed to democracy today is the use of technology, data, and automated systems in ways that threaten the rights of the American public. Too often, these tools are used to limit our opportunities and prevent our access to critical resources or services,” the White House’s announcement explains.

“These problems are well documented,” the OSTP continues. “In America and around the world, systems supposed to help with patient care have proven unsafe, ineffective, or biased. Algorithms used in hiring and credit decisions have been found to reflect and reproduce existing unwanted inequities or embed new harmful bias and discrimination. Unchecked social media data collection has been used to threaten people’s opportunities, undermine their privacy, or pervasively track their activity — often without their knowledge or consent.”

The proposed solution is, rather than trying to put the genie back in the bottle, to develop an equivalent of the Bill of Rights for artificial intelligence technology — a document built around five key principles, which the White House Office of Science and Technology claims will help keep citizens safe from rogue AI, whether misapplied or actively abused.

Those principles: that automated systems should be safe and effective, developed in consultation with diverse communities, stakeholders, and domain experts with a view to identifying and mitigating risks and negative impacts; that there should be protections in place against discrimination based on the output of algorithms; that citizens’ data should be safeguarded from abuse via protections built into the system itself; that use of automated systems should come with the requirement to provide notice and explanation to those affected by it; and that citizens should be able to opt-out “where appropriate” and deal with a person instead.

“The Blueprint for an AI Bill of Rights is a guide for a society that protects all people from these threats — and uses technologies in ways that reinforce our highest values,” the OSTP’s announcement claims. “Responding to the experiences of the American public, and informed by insights from researchers, technologists, advocates, journalists, and policymakers, this framework is accompanied by From Principles to Practice — a handbook for anyone seeking to incorporate these protections into policy and practice, including detailed steps toward actualizing these principles in the technological design process. These principles help provide guidance whenever automated systems can meaningfully impact the public’s rights, opportunities, or access to critical needs.”

The resulting framework isn’t applicable to all forms of automation. Instead, it applies only to automated systems, which have the potential to “meaningfully impact the American public’s rights, opportunities, or access to critical resources or services.”

It is also not yet legally binding: as a “blueprint,” the OSTP warns that its guidance is “non-binding” and “does not constitute US government policy” — and specifically allows for “a balancing of equities, for example, between the protection of sensitive law enforcement information and the principle of notice” when automated systems are used for the purposes of law enforcement or national security.

The OSTP’s blueprint is now available to download, along with supporting documentation, on the White House website.