President Joe Biden on Monday issued an executive order on the safe and secure use of artificial intelligence, including a provision directing the National Security Council (NSC) and his chief of staff develop a national security memorandum to guide the Defense Department and the intelligence community (IC) on the safe and ethical use of the technology for their missions.
The memorandum also would direct DoD and the IC to counter military use of AI by enemies.
“This document will ensure the United States military and intelligence community use AI safely, ethically, and effectively in their missions, and will direct actions to counter adversaries’ military use of AI,” a White House fact sheet says.
Another provision of the directive calls for the National Institute of Standards and Technology (NIST) to create standards for red-teams to test AI systems to ensure they are safe prior to public release. NIST, which is part of the Department of Commerce, creates consensus-based standards with private and public sector input.
Once safe, secure, and trustworthy standards are established, the Department of Homeland Security will apply these standards to critical infrastructure sectors and establish an AI Safety and Security Advisory Board (AISSB) to be chaired by the secretary of homeland security. DHS will work with stakeholders on guidance for critical infrastructures to use.
“The AISSB will issue recommendations and best practices for an array of AI use cases to ensure AI deployments are secure and resilient,” DHS said.
In addition, DHS and the Department of Energy will assess threats posed by AI systems to critical infrastructure, and chemical, biological, radiological, nuclear, and cybersecurity risks.
The Cybersecurity and Infrastructure Security Agency, a component of DHS, will also work with international partners to harmonize guidelines for critical infrastructure sectors to close gaps in their AI systems against physical and cyber-attacks.
In the area of cyber security, DHS said that CISA is already using AI and machine learning to detect and prevent threats, and to conduct vulnerability assessments.
“Furthermore, DHS will conduct an operational test to evaluate AI-enabled vulnerability discovery and remediation techniques for federal civilian government systems,” DHS said.
For weapons of mass destruction, DHS and the White House Office of Science and Technology Policy will examine the “potential for AI to lower the barriers to entry for developing WMD,” in particular for the technology to use synthetic nucleic acids to create biological weapons, DHS said.
The department will develop a framework and standard test expectation for third parties to audit AI systems for the misuse of synthetic-nucleic acids.
The executive order “directs DHS to manage AI in critical infrastructure and cyberspace, promote the adoption of AI safety standards globally, reduce the risk of AI’s use to create weapons of mass destruction, combat AI-related intellectual property theft, and ensure our immigration system attracts talent to develop responsible AI in the United States,” Homeland Security Secretary Alejandro Mayorkas said in a statement. “The AI Safety and Security Advisory Board…will bring together industry experts, leading academics, and government leaders to help guide the responsible development and safe deployment of AI.”
The executive order also calls for DHS to develop a plan to reduce intellectual property risks related to AI. DHS will also streamline applications for foreign nationals who want to study and work on AI and other emerging technologies in the U.S.