On October 30, 2023, President Biden signed the 53-page Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence that significantly advances the United States' policy framework regarding artificial intelligence (AI). This directive builds upon the Biden Administration’s previous Blueprint for an AI Bill of Rights and outlines a comprehensive strategy aimed at positioning the United States as a leader in the guidance of responsible AI development and application.

While recognizing the potential of responsible AI systems to make the world more prosperous, productive, innovative, and secure, it also acknowledged that irresponsible use could make societal issues such as fraud, discrimination, bias, and disinformation worse, as well as resulting in displaced and disempowered workers, stifled competition, and risks to national security.

The Executive Order also recognizes that addressing these issues requires the coordination of the government, private sector, academia, and civil society. While most of the requirements in the Executive Order apply only to the federal government, private businesses may be impacted by a few requirements that directly apply to private enterprises, requirements that may apply to businesses that contract with the federal government for the use of AI, and any resulting statutes, regulations, or guidance that are required to be developed under the Executive Order.

Guiding Principles

The Executive Order sets out eight guiding principles and priority regarding the responsible development and use of AI:

• Safe and Secure AI: AI systems must undergo robust, reliable, repeatable, and standardized evaluations to ensure their safety and security. Policies, institutions, and mechanisms should be in place to test, understand, and mitigate risks from these systems before they are used. The federal government must be notified, including the results of the safety tests, when training AI systems that may pose a risk to national security, economic stability, or public health and safety.
• Leadership by the United States. The Executive Order recognizes that by promoting responsible innovation, competition, and collaboration, the United States can lead in AI and unlock its potential. The United States should invest in education, training, R&D, and capacity to attract the best AI talent and to promote responsible innovation, competition, and collaboration. The novel intellectual property rights questions that arise from the development and use of AI should be addressed, and unlawful collusion and monopoly over key assets and technologies should be stopped. The Executive Order also states that the federal government will promote a fair, open, and competitive ecosystem and marketplace for AI and related technologies to allow small developers and entrepreneurs to drive innovation in this area and prevent unlawful collusion that could arise in semiconductors, computing power, cloud storage, and the availability and use of data.
• Support of American Workers. The Executive Order recognizes that the responsible development and use of AI requires a commitment to supporting American workers through education and job training. The impact of AI on the labor force and workers’ rights should be understood. The Executive Order therefore states that the administration will seek to adapt job training and education to prepare American workers for the deployment of AI while attempting to make sure that AI is not deployed in ways that undermine rights, worsen job quality, encourage undue worker surveillance, lessen market competition, introduce new health and safety risks, or cause harmful labor-force disruptions. This is in line with the goal of supporting responsible uses of AI that improve the American workers’ lives, positively augment their work, and help people safely enjoy the gains from this technology.
• Equity and Civil Rights. The Executive Order recognizes that AI policies must be consistent with the advancement of equity and civil rights, including addressing “algorithmic discrimination.” To further this goal, the Executive Order promises that the administration will build on steps that have already been taken by the federal government (such as the NIST AI Risk Management Framework and Executive Order No. 14091) to ensure that AI complies with all federal laws and promotes appropriate evaluation and oversight. This includes holding those responsible for developing and deploying AI accountable to standards that protect against discrimination and abuse.
• Consumer Protection. Another guiding principle outlined in the Executive Order is that the interests of Americans, especially minors, who increasingly use, interact with, or purchase AI and AI-enabled products in their daily lives must be protected, and that the use of AI cannot excuse organizations from their legal consumer protection obligations. As a result, the Executive Order promises that the federal government will continue to enforce existing consumer protection laws and enact appropriate safeguards against fraud, unintended bias, discrimination, invasions of privacy, and other similar harms related to the use of AI – especially in the areas of healthcare, financial services, education, housing, and transportation where errors can be disastrous. At the same time, the federal government will promote responsible uses of AI to protect consumers, improve the quality of goods and services, lower prices, and expand selection and availability.
• Privacy and Civil Liberties. The Executive Order outlines that Americans’ privacy and civil liberties must be protected by ensuring that the collection, use, and retention of data is lawful, secure, and promotes privacy. Recognizing that AI makes it significantly easier to extract, re-identify, link, infer, and act on sensitive information about people and their lives, the Executive Order requires that the federal government ensure that data collection, use, and retention is lawful, secure, and addresses privacy and confidentiality risks by requiring federal agencies to use available policy and technical safeguards (including privacy-enhancing technologies) to protect individual privacy and fight against legal and societal risks.
• Responsible Use by Federal Government. The Executive Order recognizes that the federal government should manage the risks from its own use of AI and increase its internal capacity to regulate, govern, and support the responsible use of AI to deliver better results for Americans. The Executive Order promises that the federal government will work to make sure that federal employees will get adequate training to understand the benefits, risks, and limitations of AI while attempting to modernize and streamline its operations.
• International Coordination. The Executive Order recognizes that the federal government should lead the way to global societal, economic, and technological progress by engaging with international partners to develop a framework to manage AI risks, unlock AI’s potential for good, and promote a common approach to shared challenges. To do this, the Executive Order promises that the administration will engage with the international community to develop a responsible framework for the use of AI.

Regulatory Requirements

The Executive Order sets forth a number of specific requirements designed to realize these principles. While much of the Executive Order creates obligations for cabinet members and agency heads, the results of many of these activities are likely to impact private businesses. Specifically, the Executive Order requires that cabinet members and agency heads, generally in cooperation with private industry, to enact policies and procedures and take other actions that may impact private businesses in the following ways:

• Setting new standards for AI safety and security, including:
  
  • Requiring that developers of powerful AI share their safety test results and other critical information with U.S. regulators.
  • Developing standards, tools, and tests to evaluate the safety, security, and trustworthiness of AI.
  • Developing standards to protect against the risk of using AI to develop dangerous biological materials.
  • Protecting against fraud and deception by establishing standards and best practices designed to detect AI generated content and authenticate legitimate content.
  • Developing AI tools to find and fix vulnerabilities in critical software.
• Calls to pass comprehensive privacy legislation to protect Americans and especially minors by:
  
  • Prioritizing federal support for privacy enhancing technologies to protect privacy, including ones that use AI and that lets AI systems be trained while preserving privacy.
  • Funding a research coordination network to strengthen privacy-preserving research and technologies.
  • Developing guidelines for federal agencies to evaluate privacy enhancing technologies, which may affect private-sector businesses that provide AI and such technologies.
• Advance equity and civil rights by:
  
  • Providing clear guidance to landlords, administrators, and federal benefits programs on the proper and responsible use of AI and how to avoid using AI in a way that would increase discrimination.
  • Addressing algorithmic discrimination through training, technical assistance, and coordination between federal agencies investigating civil rights violations related to AI.
  • Ensuring fairness throughout the criminal justice system by developing best practices on the use of AI in criminal justice activities.
• Stand up for consumers, patients, and students by:
  
  • Advancing the responsible use of AI in the health care and pharmaceutical industries.
  • Creating resources to support educators deploying AI-enabled education tools.
• Mitigate risks to workers from the dangers of increased workplace surveillance, bias, and job displacement by:
  
  • Developing principles and practices that address job displacement, labor standards, workplace equity, health, and safety, and data collection to minimize the harms of AI for workers.
  • Developing reports on the impact of AI on labor-markets and study and identify options for supporting workers who may face labor disruptions from AI.
• Promote innovation and competition by:
  
  • Piloting a tool that will provide AI researchers and students with access to AI resources and data (National AI Research Resource), and expand grants for AI research in vital areas, such as health care and climate change.
  • Providing small developers and entrepreneurs with access to technical assistance and resources and helping small businesses commercialize AI technology.
  • Enhancing the AI workforce of highly skilled immigrants and non-immigrants by modernizing and streamlining visa criteria, interviews, and reviews.
• Take a leadership role in supporting safe, secure, and trustworthy development and use of AI by:
  
  • Expanding international agreements and frameworks to collaborate on the responsible use of AI to maximize its benefits and managing its risks.
  • Accelerating the development of international standards.
  • Promoting the safe, responsible, and rights affirming development and deployment of AI
• Ensuring responsible and effective government use of AI by:
  
  • Issuing guidance on use of AI by federal agencies that includes clear standards to protect the rights of individuals and public safety while improving procurement and strengthening deployment.
  • Assisting in acquiring AI products and services quicker and cheaper.
  • Accelerating the hiring of AI professionals in the U.S. government and requiring agencies to provide AI training for government workers at all levels.

Key Takeaways

The definition and impact is seemingly broad. The Executive Order defines an “AI system” to include any data system, software, hardware, application, tool, or utility that operates in whole or in part with AI. The current definition does not narrow what specific products, software, or geographic locations are included in an “AI system.” This leaves it open to interpretation as to how far the new rules and regulations will reach.

This is the start of potentially rigorous regulation. Many agencies have been tasked with directives to develop national standards to ensure the safety and security of AI. It is possible that some agencies will develop standards or regulations that will mirror or even conflict with each other. However, until the agencies start promulgating regulations it is unclear how this will affect AI in the long-term.

The administration is not shy about their desire to promote competition. Not only has the Executive Order called for small developers and entrepreneurs to be given access to technical assistance and resources to help commercialize AI breakthroughs, but it calls for streamlining the process for noncitizens to conduct research in AI and other critical and emerging technologies. Further, a pilot of the National AI Research Resource will be launched as a tool to provide AI researchers and students access to key AI resources and data.

Concern for safety and security is at the forefront. The Executive Order focuses on protecting individuals’ rights and safety, with hopes to mitigate the risk of discrimination and bias from AI systems. Specifically, there are directions to advance equity and civil rights, reduce the potential for misleading the public, and support workers in workplace surveillance, bias and job displacement.

Looking Ahead

Much of the Executive Order directs cabinet members and heads of agencies to develop guidance or regulations regarding the use of AI that has yet to be developed. That being said, it also imposes deadlines on federal agencies to issue reports and draft guidelines that address key concerns echoed by the administration. If Congress enacts further legislation in response to the Executive Order, it is vital for businesses to be attentive to these new rules and take necessary steps to comply.

Companies should ensure they have assistance from counsel to help guide them through this increasingly complex legal landscape. Continuing to monitor for such guidance or regulations and preparing to implement them as appropriate when they are finally released will go a long way to maximizing the benefits of AI while minimizing disruption to operations. In the meantime, organizations may wish to begin self-audits and monitor systems currently under development or already deployed for potential data privacy and cybersecurity risks, erroneous outputs, and bias.

Mackenzie N. Barrett and Robin R. Zhang contributed to this article.