The White House Office Of Science And Technology Policy Reveals AI Bill Of Rights
Today the White House Office of Science and Technology Policy revealed the AI Bill of Rights, a set of principles for the safe and equitable use of AI on the American people.
Created on October 4|Last edited on October 4
Comment
With the increasingly rapid widespread adoption of AI across countries and industries, paying mind to the ways that AI can harm individuals when applied maliciously or poorly is important.
Today, the White House Office of Science and Technology Policy released the blueprint for the AI Bill of Rights, with the aim of setting a guideline for the development and application of AI in areas where it might present problems in privacy and equity.
The AI Bill of Rights presents 5 key protections they believe everyone in America should be entitled to, covering topics of personal data privacy, discrimination mitigation, and safe systems, as well as notice that AI systems are being used and the option to opt-out where applicable.
The 5 Key Protections Of The AI Bill Of Rights Are
- Safe and Effective Systems - AI systems should be developed with diverse consultation, safety testing, and without irrelevant data use.
- Notice and Explanation - People should be notified when they are subject to automated systems, including how it contributes to outcomes.
- Human Alternatives, Consideration, and Fallback - People should be able to opt-out of AI systems in favor of a human alternative.
These protections are "common sense" as they put it in the OSTP blog post, and that's certainly true for most people engaging with the AI industry as it grows.
Many of the bigger-named companies pushing AI forward every day have their own development principles they follow, and are usually quite vocal about.
The US government's statements today put emphasis on the average person's side of the equation - making sure that AI models are used responsibly for the good of the average person.
Keep in mind that this AI Bill of Rights is for "automated systems that have the potential to meaningfully impact the American public's rights, opportunities, or access to critical resources or services", so unless your AI model is geared up to be implemented into something like employment or banking, you're unaffected by any of this.
Find out more
Add a comment
Tags: ML News
Iterate on AI agents and models faster. Try Weights & Biases today.