Today, 30 October 2023, U.S. President Joe Biden's administration signed an Executive Order on artificial intelligence to establish AI safety and security standards to protect Americans’ privacy and civil rights.
AI safety and security
The Executive Order takes decisive action against AI, with sweeping measures for powerful AI systems to protect citizens from the risks of the technology.
In particular, the Executive order:
- Requires developers of foundational AI models to share results of safety tests with the U.S. Government in accordance with the Defense Production Act if they pose a serious risk to national security, national economic security, or national public health and safety.
- Requires the National Institute of Standards and Technology to set strict standards for red-team testing of powerful AI models before they can be released.
- Requires agencies that fund life-science projects to establish standards for biological synthesis screening to prevent AI being used to engineer dangerous biological materials.
- Will establish an advanced cybersecurity program to develop AI tools to identify and correct vulnerabilities in critical software to enhance security.
- Directs the development of a National Security Memorandum that directs additional actions on AI and security to ensure that the U.S. military and intelligence use AI safely, ethically, and effectively.
As well as protecting against AI misuse, the Executive Order will also impose requirements to protect the privacy of American Citizens, with the President calling on Congress to pass data privacy legislation at the federal level. This legislation is envisioned to:
- Prioritise federal support for accelerating the development and use of privacy-preserving techniques.
- Strengthen investment and research in privacy-preserving tools and technologies and promote their adoption of federal agencies.
- Provide oversight into how agencies collect and use commercially available information and strengthen federal agency privacy guidance to manage the use of personally identifiable data.
- Develop guidelines for the evaluation of the effectiveness of privacy-preserving techniques for federal agencies.
Promoting equity and protecting civil rights
The Executive Order particularly focuses on protecting the equal opportunity and non-discrimination rights of U.S. citizens, requiring that:
- Clear guidance is given to landlords, Federal benefits programs, and Federal contractors to prevent algorithmic discrimination.
- There is training, technical assistance, and coordination between the Department of Justice and Federal civil rights offices to establish best practices for handling civil rights violations resulting from AI.
- Best practices are developed for the use of AI in criminal justice activities such as sentencing, parole, probation, pretrial released and detention, risk assessments, and surveillance.
Protection for consumers, patients, and students
The Executive Order particularly focuses on protecting consumers, patients, and students to balance innovation with harm prevention. In particular, the Executive order seeks to protect these groups by:
- Advancing the responsible use of AI in healthcare and drug development and establishing a safety program for reports and mitigation of harms or unsafe AI applications in healthcare.
- Create resources to shape the transformative potential of AI in education through initiatives such as personalised tutoring.
Support for workers
Given that the transformative power of AI has the potential to displace workers, the Biden Administration directs the following:
- Address job displacement through the development of principles and best practices to mitigate harms and maximise benefits of AI for workers by providing guidance against unfair compensation, unfair evaluation of job applicants, or impingement on workers’ ability to organise.
- Study and identify options for federal support for displaced workers and produce a report on potential impacts on the labour market.
Promotion of innovation and competition
The Biden Administration seeks to maintain the U.S.’s position as a global leader in AI innovation by:
- Piloting a National AI Research Resource to provide AI researchers and students access to key AI resources and data as well as grants for vital AI research.
- Providing small developers and entrepreneurs access to technical assistance and resources to promote a fair AI competition ecosystem.
- Expanding the ability of highly skilled immigrants and nonimmigrants with expertise in critical domains to work and study in the U.S. through changes to visa processes.
Advancement of American leadership abroad
As well as focusing on the U.S.’s capabilities, the Executive Order also directs actions for international cooperation towards safe, secure and trustworthy AI:
- Expand stakeholder engagements on AI to establish international frameworks for AI risk management.
- Accelerate the development and implementation of AI standards in collaboration with international partners and standards organisations.
- Promote the adoption of appropriate safeguards abroad to promote the safe, responsible and rights-affirming development and deployment of AI globally.
A focus on government use of AI
A key focus of the Executive Order is government and Federal agency assessment and use of AI, with the Presidential decree directing the following to promote the responsible deployment of AI by Federal agencies, as well as to modernise Federal infrastructure:
- Issue guidance and clear standards for Federal agency use of AI to protect rights and safety, improve procurement, and strengthen deployment.
- Assist agencies in acquiring specified AI products and services more efficiently through changes to contracting.
- Accelerate the rapid hiring of AI professionals in the government as part of an AI talent surge that will see agencies provide AI training for employees at all levels.
Other U.S. Federal efforts
The Executive Order follows other recent efforts towards responsible AI by the Biden Administration, including the Blueprint for the AI Bill of Rights published by the White House in 2022 and voluntary commitments from tech giants to manage the risks of AI.
Today’s new Executive Order is likely to have the most significant implications for Big Tech, with companies such as Amazon, Microsoft, and Google soon being required to disclose their safety testing procedures and results to the U.S. Federal government.
While the impact of the order remains to be seen, this could be an important step towards AI transparency.
Holistic AI are pioneers in the field of AI Governance, Risk, and Compliance. We specialise in giving organisations the tools to comply with AI legislation in the U.S. and beyond, helping them to harness AI’s power effectively and responsibly.
Schedule a call with one of our compliance experts to discover how we can help your business.