7 March 2021

A Solid Plan to Compete with China in Artificial Intelligence

By MEGAN LAMBERTH and MARTIJN RASSER

If the United States is to keep ahead of a rapidly gaining China in the field of artificial intelligence, it needs a concrete and comprehensive plan for action. Such a plan is presented in the final report, released today, of the National Security Commission on Artificial Intelligence, or NSCAI. Critically, this report is about more than AI. It is the opening salvo of a much-needed effort to create an overarching national strategy for technology, a whole-of-government effort to safeguard American technological leadership.

Congress created the NSCAI three years ago to determine how the United States could develop AI and machine learning systems to address U.S. national security and defense needs. The Commission’s recommendations, and the urgency it conveys, are likely to shape the U.S. government’s AI strategy in the years to come, particularly within the Defense Department. The report makes clear that U.S. supremacy in AI is not a given, and that the government must act swiftly and effectively to harness the technology’s transformative power.

At more than 700 pages, the report is one of the most comprehensive documents on AI competitiveness ever written. Its first half describes how the U.S. can adopt AI to “change the way we defend America, deter adversaries, use intelligence to make sense of the world, and fight and win wars.” Among the myriad of recommendations, two sets stand out. One is a call for an “AI-Ready DoD” by 2025, which amounts to a comprehensive overhaul of the Department’s approach to AI. It spans digital literacy and infrastructure to crafting new concepts and operations to integrate AI technologies. The underlying admonition is that the U.S. military is at serious risk of failing to effectively wield AI-enabled technologies despite enjoying access to world-leading capabilities.

The second set is likely to prove the most controversial of the entire report, as it concerns the lawfulness, safety, and ethics of AI-enabled and autonomous weapons. In a series of judgments, the NSCAI concludes that such weapons have been and can continue to be used in accordance with international humanitarian law, and that DoD has the procedures in place to ensure this remains the case. As a result, the Commission does not support a global prohibition of such weapons. The associated recommendations focus on mitigating the strategic risks of AI, including ensuring human decision making for the use of nuclear weapons, crisis stability dialogues with Russia and China, and setting international standards for the development, testing, and use of AI-enabled and autonomous weapons. The Commission is unambiguous that AI-enabled warfare is the future.

The report’s second section—“Winning the Technology Competition”—explains how AI fits into the broader strategic competition between the U.S. and China. While many of the recommendations in this section have been advocated for in the Commission’s interim reports, some of the specifics have changed. Notably, the NSCAI now calls for even greater R&D investments—at least $8 billion a year by the Defense Department, federal funding of AI R&D at $32 billion annually by 2026, and federal spending of 1 percent of GDP for R&D for science across the board. This is the boldest pronouncement yet among a growing chorus of support for increased AI R&D spending.

An important recommendation is to create a Technology Competitiveness Council, a proposed bureaucratic entity within the White House that would unify the various technology policy efforts underway in the executive branch. Led by a new principal, the council would be responsible for crafting and executing a national technology strategy, with an initial focus on the AI talent competition, fostering innovation, protecting U.S. competitive advantages, and partnering with like-minded countries to create a “favorable international technology order.” Given the centrality of technology to the economic vitality and national security of the United States, the NSCAI is making a profound statement on the need for greater leadership by the federal government if it seeks to ensure long term U.S. competitiveness.

Many of the report’s recommendations align with the Biden administration’s key policy priorities, including on supply chain security, talent, and the stakes of the U.S.-China competition. The Commission warns that the U.S. is “overly dependent upon globally diversified supply chains for microelectronics”—language echoing the Biden administration’s recent executive order on supply chain security. The Commission declares immigration reform a “national security imperative” and examines ways the government could create new immigration pathways and bolster existing ones—recommendations that may help shape the Biden administration’s immigration agenda. The Commission is unequivocal in describing the high stakes of the technology competition between the U.S. and China, of which AI is a critical component.

The Commission’s message is clear: the U.S. is still ahead of China in AI, but that gap is shrinking rapidly. Their report offers a concrete and comprehensive plan for action that should resonate with many in the White House and on Capitol Hill. The stakes couldn’t be higher. NSCAI did its job with its final report. Now America’s policymakers must act.

Megan Lamberth is a Research Associate for the Technology and National Security Program at the Center for a New American Security (CNAS). Martijn Rasser is a Senior Fellow with the CNAS Technology and National Security Program. They are the authors of the CNAS report, “Taking the Helm: A National Technology Strategy To Meet the China Challenge.”

No comments: