Earlier this month, the White House Office of Science and Technology released its ‘Blueprint for an AI Bill of Rights’. This bill has been created as a guideline to mould how AI systems are designed, developed and deployed, and is intended to be a “guide for a society that protects all people from…threats and uses technologies in ways that reinforce our highest values.”

The bill has been developed with the goal of reducing and eliminating the negative impacts that automated systems have on our lives and increasing their potential positive effects and applications.

The White House names certain effects such as “unsafe, ineffective, or biased” systems being used in patient care, biased algorithms being adopted in hiring practices and for credit decisions, and biased algorithms in general reflecting and reproducing real world bias and discrimination as examples of what the bill hopes to tackle.

AI has had, and will continue to have a transformative effect on our lives, so bills such as this are very important to ensure those effects are overwhelmingly positive. Furthermore, seeing as America is home to some of the largest and most powerful tech and AI companies, it is particularly important that the White House takes this kind of action.

“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.” 

– The White House

One of the main factors of the bill is the proposal of five core principles identified by the White House Science and Technology Policy as central issues to tackle in order to ensure the utmost protection against the misuse of AI.

These principles include:

Safe and effective systems – you should be protected from unsafe or ineffective systems.

Algorithmic discrimination protection – you should not face discrimination by algorithims and systems should be used and designed in an equitable way.

Data Privacy – you should be protected from abusive data practices via built-in protections and you should have agency over how data about you is used.

Notice and explanation – you should know that an automated system is being used and understand how and why is contributes to outcomes that impact you

Human alternatives, consideration, and fallback – you should be able to opt out, where appropriate, and have access to a person who can quickly consider and remedy problems you encounter.

The bill also acknowledges the definitional problems that arise with AI – it can be hard for regulation to keep up with the ever-occurring developments in this sector, so alongside these core principles, they have created a “two-part test to determine what systems are in scope.”

This includes stating that the framework applies to:

“(1) automated systems that (2) have the potential to meaningfully impact the American public’s rights, opportunities, or access to critical resources or services”, and that the “framework described protections that should be applied with respect to all automated systems that have the potential to meaningfully impact individuals’ or communities’ exercise of…rights, opportunities, or access”.

Whilst what is set out in the bill seems to be a promising blueprint for the regulation and improved accountability of AI, its main criticism is that the principles and advice it sets our are non-enforceable and quite generalized, and it does not provide much in terms of explicit grounds for regulation. However, here at DataEthics4All, we recognise that, when misused, the applications of AI can pose threats to our autonomy in general, and cause disproportionate harm to minority groups. So, we are in support of the new AI Bill of Rights and hope that, despite its criticisms, if it is enforced correctly and followed up meaningful regulation, that it will meet its goals of making AI more accountable.

Relevant Links:

https://www.whitehouse.gov/ostp/ai-bill-of-rights/

Contact connect@dataethics4all.org to learn more.