President Joe Biden has introduced new initiatives to leverage Artificial Intelligence (AI) for national security amid the rapidly advancing global race to develop the technology. On Thursday, Biden presented the inaugural AI-specific National Security Memorandum (NSM), emphasizing the necessity for the government to lead in the creation of “safe, secure, and trustworthy” AI. The memorandum instructs US agencies to strengthen their semiconductor chip supply chains, integrate AI considerations into emerging government technology, and prioritize intelligence efforts to counter foreign attempts that threaten US AI dominance. “We must out-compete our adversaries while addressing the threats posed by their use of AI, ” remarked a Biden administration official as reported by AFP, the French news agency. The White House memo highlighted the importance of using AI to uphold human rights and democratic principles. It noted, “Americans need to be assured that they can trust systems to operate safely and reliably. ” To enhance protective measures, the memo mandates US agencies to “monitor, assess, and mitigate AI risks associated with privacy invasions, bias and discrimination, individual and group safety, and other human rights violations. ” Additionally, it calls for a framework that enables Washington to collaborate with allies to ensure AI is developed and utilized in compliance with international law while safeguarding human rights and fundamental freedoms. This memorandum is the latest effort from Biden’s administration to address the rapidly evolving technology, which US officials anticipate will provoke heightened military and intelligence rivalry among global powers. Last year, Biden enacted an executive order aimed at mitigating AI-related risks to consumers, workers, minority communities, and national security. However, in July, over a dozen civil society organizations, including the Center for Democracy & Technology, sent an open letter to Biden administration officials advocating for strong safeguards to be embedded in the NSM.

They noted that “little is known” about the use of AI by government agencies despite their “commitments to transparency. ” They expressed concern that AI’s application in national security contexts might reinforce racial, ethnic, or religious biases while exacerbating violations of privacy, civil rights, and civil liberties. Next month, the United States will host a global safety summit focused on AI in San Francisco, where allies will collaborate on improving regulations in the sector and coordinating policies. Generative AI, which can produce text, images, and videos based on open-ended prompts, has sparked both excitement about its possibilities and concerns regarding its potential misuse and the risk of overshadowing human capabilities with dire consequences.