Everything You Need To Know About The New Executive Order
Today, the Biden Administration signed an Executive Order on the “safe, secure, and trustworthy” use of AI. The sweeping order—extensive in both the topics covered and organizations involved—builds on previous White House efforts and sets out actions to establish new standards for AI safety. The order looks to protect the rights of Americans, combat AI discrimination, encourage the use of safe AI in government, and more.
This order is the most significant federal effort to date to regulate the fast-moving technology and will shape future efforts to regulate AI in the US.
So what does this mean for you? Let’s break down the most important takeaways for the corporate world.
Details of the Order
The Order sets out many actions across different topics.
New Standards for AI Safety and Security
The first set of requirements is aimed at safety and transparency for cutting-edge frontier models, which have seen the most astonishing increase in capabilities, but also pose some of the most serious threats.
Developers of foundation models that pose a risk to national security, economic security, or public health, and safety will be required to inform the government of their work, including safety and red-teaming results.
The order also calls for the development of content authentication and watermarking techniques (commonly referred to as “watermarking”) for AI-generated content. The goal of these actions is to ensure foundational models are safe before they reach the public.
Additional actions are taken to prevent the use of AI in creating biological weapons and to develop AI tools that can identify and address cybersecurity vulnerabilities.
Equity and Civil Rights
A key part of the White House’s AI Bill of Rights, the order recognizes the advancement and protection of civil rights as a critical way to mitigate the potential pitfalls of AI.
The order provides guidance to landlords, federal benefits programs, and federal contractors to prevent discrimination by AI. It also calls for training, technical assistance, and coordination between the Department of Justice and federal civil rights offices to establish best practices for investigating and prosecuting civil rights violations involving AI. Lastly, it seeks the development of best practices for using AI in the criminal justice system.
Responsible Use of AI in Government
Similar to the private sector, AI can enhance government services for the public, but it also entails corresponding risks. To address this, the order provides guidance to government agencies to safeguard rights and ensure safety. It also streamlines the procurement of AI products and services, expediting the hiring of AI professionals to enhance efficiency.
Privacy Protections, and Protections for Consumers, Patients, and Students
Acknowledging the significance of consumer data in AI systems, the order calls for measures to protect Americans’ privacy. It offers federal support for privacy-preserving techniques—methods enabling developers to train AI models using data while respecting users’ privacy. Additionally, the government will assess how agencies collect and use commercially available information, strengthening privacy safeguards.
Concerned about AI potentially automating jobs and jeopardizing workers’ rights, the order aims to establish a set of best practices to maximize benefits while minimizing harm to workers. It also mandates a report on AI’s labor market impact, exploring potential measures to support displaced workers.
To ensure AI enhances consumers’ lives, the order also promotes AI advancement for the development of affordable drugs and educational transformation.
Promoting Innovation
Given that AI presents a double-edged sword, encompassing both advantages and risks, the Administration recognizes the imperative of fostering the secure advancement of AI. To accomplish this, the order will provide essential technical resources—funds, computational assets, and technical support—to empower small developers within the ecosystem.
Additionally, it will streamline the visa process to enhance the capacity of crucial AI talents to pursue studies and employment in the United States.
Promoting American Leadership
The Administration is committed to working with allies to develop a strong international framework for the safe development and use of AI.
The Administration mentions they have already discussed framework alignment with many other countries and organizations.
The order comes two days before the UK’s AI Safety Summit and as the European Union is finalizing the details of the EU AI Act. The order, and the timing, sends a clear message that American views on AI regulation align with other countries’ perspectives, and that the country is able and willing to be one of the first to pass sweeping legislation.
Other US Regulation Efforts
The Executive Order builds on previous White House efforts, such as the AI Bill of Rights released in October, 2022 and the voluntary commitments to safe AI it secured from leading tech companies.
Compared with the EU’s AI Act or Canada’s Artificial Intelligence and Data Act, most AI legislation passed in the US has been narrow and sector-specific. Consider New York City Local Law 144, which entered enforcement this July. The bill was one of the first in the US to curb irresponsible AI usage, but it only applied to hiring and promotions in NYC. A bill called the Algorithmic Accountability Act would require algorithmic “impact assessment” of high-risk AI systems nationally. However, the bill has entered Congress three times, failing twice before being reintroduced recently in 2023, and its chances for passing are unclear.
As an Executive Order, the text lacks the legitimacy of congressional legislation and could be voided by a future president. However, the order clearly lays out the Biden Administration’s priorities and could influence a bill passed through Congress. Chuck Schumer, a senator from New York, has stated he wants to submit comprehensive legislation around AI regulation and has been in constant contact with the Biden White House. The legacy of the Executive Order may not be the order itself but how it influences US (and international) regulation.
Implications of the Order
While many of the specifics of the order have not been defined yet and the mechanisms for enforcement are uncertain, the Biden Administration makes its priorities on AI clear.
The order recognizes that AI is a fast-moving technology with both tremendous upside and downside – on the economy, civil rights, and more. It understands the importance of striking a balance between the upsides and risks. It understands the drivers of those harms and benefits.
It recognizes that regulation needs to move fast before AI outpaces our means of regulating it. And it’s committed to protecting Americans against the harms of unsafe AI.
Top 5 Takeaways for the Water Cooler Conversation
1. The Executive Order represents the most significant federal effort to date to regulate AI technology in the United States. This signals a strong commitment from the Biden Administration to ensure AI benefits are balanced with safeguards against potential harms.
2. The order emphasizes the importance of civil rights and equity in AI development and usage.
3. Recognizing the significance of consumer privacy, professionals should also consider the privacy of employee data. HR systems that leverage AI should incorporate privacy-preserving techniques to protect employee information.
4. The commitment to establishing a global framework for AI safety and development emphasizes the importance of staying updated with international AI regulations. CHROs should be aware of global AI trends and regulations, especially if their organization operates internationally.
5. Any organization building, buying, or using AI would be wise to pay attention to the Biden Administration’s message and embed responsible AI framework into their overall governance strategy.
We think this Executive Order is a step in the right direction to establishing effective AI regulation in the US, and aligns with our mission at FairNow to help companies use AI safely and responsibly.
Are you wondering how this order and related regulations will impact your business? As the leading provider of AI Governance software – you could say that this is one of our favorite topics! Let us know a little more about our organization and how we can help you take control of your AI.