Download PDF

After observing that “Artificial intelligence offers enormous promise and great risk,” the Biden administration recently announced an Executive Order (the “Order”) setting the table for future government action in this fledging area. The Order primarily focuses on safety and security requirements but also includes provisions aimed at promoting the development of AI within the United States. Seven leading AI companies—Amazon, Anthropic, Google, Inflection, Meta, Microsoft, and OpenAI—have already pledged their support for the Order. Guided by the stated principles of safety, security, and transparency, the Order purports to pave the way for responsible AI innovation. The Order focuses on the following.

• New Standards for AI Safety and Security: The Order outlines measures to safeguard against potential AI risks. It mandates that developers of powerful AI systems share their safety test results with the U.S. government, particularly for models with significant national security, economic security, or public safety implications. It sets standards, testing protocols, and safeguards for AI systems, with a stated focus on critical infrastructure and safety. Additionally, the Order emphasizes the prevention of AI-enabled risks, such as biological threats and fraud, by creating robust screening standards and authentication measures. Recognizing the unique cybersecurity risk presented by AI, the Order also establishes an advanced cybersecurity program to enhance software and network security. To ensure responsible military and intelligence use of AI and counter adversarial AI use, a National Security Memorandum is to be developed by the National Security Council and White House Chief of Staff.
• Privacy: To address privacy risks, particularly for children, the Order urges Congress to enact bipartisan data privacy legislation. It also directs (1) the prioritizing of federal support for advancing privacy-preserving techniques, (2) the strengthening of privacy-preserving research and technologies, (3) the evaluation of how agencies collect and use commercially available information, and (4) the development of guidelines for agencies to assess the effectiveness of privacy-preserving methods, particularly within AI systems. These measures collectively aim to fortify privacy protections and mitigate the risks associated with AI’s impact on personal data.
• Equity and Civil Rights: Stating that AI usage can perpetuate discrimination, bias, and injustices in various sectors, the Order directs several measures, including (1) providing clear guidance to landlords, Federal benefits programs, and federal contractors, (2) addressing and preventing algorithmic discrimination, and (3) striving for fairness throughout the criminal justice system.
• Consumer, Patient, and Student Rights: To safeguard consumers, the Order speaks to healthcare and drug development, and seeks to create resources to support educators deploying AI-enabled tools. Of note, the Order calls for guidance related to content authentication and labeling of AI content, referred to as “synthetic content.”
• Worker Rights: The Order cites concerns about workplace surveillance, bias, and job displacement. To address these risks, the Order calls for support of collective bargaining for workers, and investment in accessible workforce training and development. The Order directs the development of principles and best practices to mitigate the harms and maximize the benefits of AI for workers. It also calls for a report assessing AI’s potential impacts on the labor market, with a focus on strengthening federal support for workers facing labor disruptions, including those caused by AI.
• Innovation and Competition: The Order claims to reinforce America’s leadership in AI innovation and competition. It states that this will be achieved by catalyzing AI research across the nation through the National AI Research Resource pilot, expanding research grants in crucial areas like healthcare and climate change, and fostering a fair and competitive AI environment. It emphasizes supporting small developers and entrepreneurs with access to technical assistance and resources, aiding small businesses in commercializing AI breakthroughs, and urging the Federal Trade Commission to utilize its authorities effectively. Additionally, the Order seeks to enhance the accessibility for highly skilled immigrants and nonimmigrants in critical fields to study, work, and remain in the United States by modernizing visa criteria and review processes.
• American Leadership Abroad: The Order calls for an expansion of bilateral, multilateral, and multistakeholder engagements to foster international cooperation in harnessing the benefits of AI while managing associated risks. The State Department, in conjunction with the Commerce Department, will lead efforts to establish robust international frameworks. Secondly, the acceleration of AI standards development with international partners and standards organizations will ensure that AI technology is safe, secure, trustworthy, and interoperable. Lastly, the promotion of responsible AI development and deployment worldwide will address global challenges such as sustainable development and safeguarding critical infrastructure.
• Ensuring Responsible and Effective Government Use of AI: The Order states that AI has the potential to improve government operations, enhance regulation, benefits distribution, and cybersecurity while reducing costs. To ensure the responsible use of AI in government, however, and to modernize federal AI infrastructure, the Order directs (1) issuing guidance for agencies to set clear standards for AI usage, safeguarding rights and safety, improving procurement, and strengthening deployment, (2) expediting agencies’ acquisition of specified AI products and services through more efficient contracting processes, and (3) accelerating of hiring AI professionals as part of a government-wide AI talent surge, and agencies will offer AI training to employees across relevant fields.

By mandating that each government agency conduct a comprehensive assessment of the applicability of AI within their respective areas of policy and regulation, the United States is substantially advancing the adoption of a sector-specific framework for AI governance. This directive demonstrates a desire to address critical issues such as data privacy and strongly underscores the necessity for legislative action by Congress.

During the process of establishing priorities, delineating principles and best practices, constructing frameworks, and providing guidance within the scope of federal AI governance, it is imperative to recognize that a substantial portion of the enforceable provisions of this Order necessitates the promulgation of regulations and the initiation of administrative procedures. These measures are time-intensive, subject to judicial scrutiny, and remain susceptible to potential revocation by subsequent administrations.

Of note is the fact that the Order’s focus on security mandates and the resulting directive were fashioned by invoking the Defense Production Act of 1950, which the federal government uses in what President Biden called “the most urgent moments.”

While rules, regulations, and possible statutes are likely to follow the Order, this is just one initial step in the federal government wading into the frontier of generative AI. This Order is as much, or possibly more, about setting policy goals and direction as it is creating firm obligations with well defined boundaries. We will continue to monitor this evolving area of the law and report on significant developments, of which there will be legion.

This content was generated in part by, and prepared for publication with the assistance of, ChatGPT.