US takes action on AI: A look at Biden's executive order

In a bid to maintain the US’s lead in AI innovation, president Joe Biden's executive order aims to further catalyse AI research across the country

US President Joe Biden on October 30 signed an executive order aimed at positioning the North American country as a global leader in harnessing the potential of artificial intelligence (AI) while addressing its risks. This comes just days before the British government hosts an international summit meeting on AI safety, scheduled for November 1 and 2. Biden's executive order, invoking the Defense Production Act previously used during the Covid-19 pandemic, encompasses a wide range of objectives. Here's a look at what this executive order entails:

The executive order mandates that developers of the most powerful AI systems must share their safety test results and critical information with the US government. Specifically, it requires companies working on foundational AI models posing serious risks to national and economic security, or public health and safety to notify the government during model training and share the results of red-team safety tests.

A red team is a group of people who assess the security and safety of a system by simulating the actions of malicious actors.

The National Institute of Standards and Technology will be tasked with developing rigorous standards for extensive red-team testing to guarantee AI system safety before public release.

Furthermore, the Department of Homeland Security will apply these standards to critical infrastructure sectors and set up the AI Safety and Security Board.

To prevent the potential misuse of AI in creating hazardous biological materials, robust screening standards for biological synthesis will be developed. Life science projects funded by government agencies must follow these standards.

The executive order seeks to protect Americans from AI-enabled fraud and deception by establishing standards and best practices for detecting AI-generated content and authenticating official content.

The Department of Commerce will develop guidance for content authentication and watermarking to identify AI-generated content. Federal agencies will also adopt these tools to assure the authenticity of government communications.

An advanced cybersecurity programme will be established to develop AI tools for identifying and rectifying vulnerabilities in critical software, making software and networks more secure against cyber threats.

A National Security Memorandum will be developed to guide AI and security actions, ensuring that the U.S. military and intelligence community use AI ethically and effectively while countering adversaries' military use of AI.

To safeguard citizens’ privacy, the Biden administration has called on Congress to pass bipartisan data privacy legislation. Furthermore, federal support will be prioritised for accelerating the development and use of privacy-preserving techniques, including AI-powered approaches that protect the privacy of training data. Funding will also be allocated to support privacy-preserving research and technologies.

Agencies' collection and use of commercially available information, particularly data containing personally identifiable information, will be evaluated to address AI-related privacy risks. Guidelines will also be established for agencies to assess the effectiveness of privacy-preserving techniques in AI systems.

The administration is committed to combating algorithmic discrimination. Clear guidance will be provided to landlords, federal benefits programmes, and federal contractors to prevent AI algorithms from exacerbating discrimination.

The Department of Justice will collaborate with federal civil rights offices to establish best practices for investigating and prosecuting civil rights violations related to AI. Efforts will also be made to ensure fairness in the criminal justice system by developing best practices for AI use in various aspects, including sentencing, parole, pretrial release, and surveillance.

The order aims to protect consumers while reaping the benefits of AI in healthcare and education. A safety program for AI in healthcare will be established, and resources will be provided to support educators in deploying AI-enabled educational tools.

With AI transforming the job market, measures will be taken to mitigate the associated risks. This includes developing principles and best practices for workers to address issues like job displacement, labour standards, workplace equity, health, safety, and data collection. A report will also be produced to assess AI's impact on the labour market and identify ways to support workers facing disruptions due to AI.

In a bid to maintain the US’ lead in AI innovation, the executive order aims to further catalyse AI research across the country. This includes the pilot of the National AI Research Resource, expanding grants for AI research in crucial areas. Access to technical assistance and resources will be provided to small developers and entrepreneurs, aiding them in commercialising AI innovations.

Additionally, efforts will be made to streamline visa criteria, interviews, and reviews to attract highly skilled immigrants and nonimmigrants with expertise in critical AI areas.

Recognising the global nature of AI challenges and opportunities, the administration will collaborate with other countries to support safe, secure, and trustworthy AI deployment worldwide. This includes expanding bilateral and multilateral engagements to establish international frameworks, accelerate AI standards development, and promote responsible AI development abroad.

The executive order acknowledges the potential of AI in government operations. It encourages the responsible use of AI in government to improve efficiency, reduce costs, and enhance security. This includes issuing guidelines for agencies, efficient procurement of AI products and services, and a surge in hiring AI professionals across government agencies.