White House proposes fresh Bill of Rights to limit AI threats

The White House has unveiled a blueprint for an artificial intelligence (AI) Bill of Rights, hoping to produce a framework that will inform future development practices to protect the public from harm.

The White House Office of Science and Technology Policy has identified five principles that should guide the design, use, and deployment of automated systems to protect the American public in the age of artificial intelligence.

The blueprint aims to act as a guide for society, providing suggestions on how people should be protected from threats while using the technology to reinforce its highest values.

It stated that citizens should be protected from unsafe or ineffective systems. Automated systems should be developed with consultation from diverse communities, stakeholders, and domain experts to identify concerns, risks, and potential impacts of the system.

They should also undergo pre-deployment testing and ongoing monitoring to demonstrate they’re safe and effective, it was proposed.

Secondly, the blueprint highlighted that citizens shouldn’t face discrimination by algorithms and systems should be used and designed in an equitable way.

It calls on designers and developers of automated systems to take proactive and continuous measures to protect individuals and communities from algorithmic discrimination and to use and design systems responsibly.

The blueprint also stated that citizens should be protected from abusive data practices via built-in protections and they should have agency over how their data is used.

“You should be protected from violations of privacy through design choices that ensure such protections are included by default, including ensuring that data collection conforms to reasonable expectations and that only data strictly necessary for the specific context is collected,” the blueprint stated. It called on designers and developers of the systems to seek permission and respect when it comes to the collection, use, access, transfer, and deletion of user data in appropriate ways.

Individuals and their communities should also be free from unchecked surveillance, it proposed. These technologies should be subject to heightened oversight which includes at least a pre-deployment assessment of their potential harms and scope limits to protect privacy and civil liberties.

Continuous surveillance and monitoring should not be used in education, work, housing, or in other contexts where the use of such surveillance technologies is likely to limit rights, opportunities, or access, read the blueprint.

Citizens should also know when an automated system is being used and understand how and why it contributes to outcomes that impact them.

Designers and developers should provide plain-language documentation including clear descriptions of the overall system functioning and the role automation plays, a notice that such systems are in use, what individual or organisation is responsible for the system, and explanations of outcomes that are clear, timely, and accessible.

Lastly, citizens should be able to opt out, where appropriate, and have access to a dedicated individual who can quickly consider and remediate problems they encounter.

“You should be able to opt out from automated systems in favour of a human alternative, where appropriate,” read the blueprint. “Appropriateness should be determined based on reasonable expectations in a given context and with a focus on ensuring broad accessibility and protecting the public from especially harmful impacts.”

Despite the Biden administration pushing forward this framework for a potential AI bill in the future, the US doesn’t have a comprehensive law to regulate the technology. It comes at a time when the European Union proposed in April a set of wide-reaching regulations to combat the misuse of AI.

It aims to protect citizens from the use of AI for mass surveillance by law enforcement and class AI in recruitment or border control management as high risk due to discrimination concerns.

In its current state, the proposed guidelines in the blueprint would not be binding, meaning that if they were formally adopted, they would not be passed into law and would merely act as optional guidelines - a weaker approach than the EU's.