Final Recommendations from the National Security Commission on AI
The global technology competition between the United States and the People’s Republic of China becomes more evident by the day.
President Xi Jinping is expected to call for a new 5-year plan this week for the PRC to achieve technology independence from the West at an annual meeting of the legislature.
On Monday, the Biden Administration released a report detailing the President’s 2021 trade agenda and countering China will be a top focus. “China’s coercive and unfair trade practices harm American workers, threaten our technological edge, weaken our supply chain resiliency, and undermine our national interests,” the U.S. Trade Representative wrote, “Addressing the China challenge will require a comprehensive strategy and more systematic approach than the piecemeal approach of the recent past.” The USTR report explains that the administration is committing to a comprehensive review and will use “all available tools to take on the range of China’s unfair trade practices” and “address the widespread human rights abuses of the Chinese Government’s forced labor program.”
Key recommendations from the National Security Commission on AI’s final report
As the Biden administration conducts this review, a new source of insightful bipartisan recommendations is the National Security Commission on Artificial Intelligence’s final report issued Monday. The Commission’s report opens by providing a helpful analogy for how to think about AI and its potential to change the world:
In 1901, Thomas Edison was asked to predict electricity’s impact on humanity. Two decades after the development of the light bulb, he foresaw a general-purpose technology of unlimited possibilities. “[Electricity] is the field of fields,” he said. “It holds the secrets which will reorganize the life of the world.”2 AI is a very different kind of general-purpose technology, but we are standing at a similar juncture and see a similarly wide-ranging impact. The rapidly improving ability of computer systems to solve problems and to perform tasks that would otherwise require human intelligence is transforming many aspects of human life and every field of science. It will be incorporated into virtually all future technology. The entire innovation base supporting our economy and security will leverage AI. How this “field of fields” is used—for good and for ill—will reorganize the world.
Last week, Eric Schmidt previewed the Commission’s broad recommendations in Congressional testimony. The 756-page report rightly goes into much greater detail about the challenges related to “defending America in the AI era” and “winning the technology competition.” The following is an overview of some of the notable and specific recommendations based on my initial review of the report, particularly focused on those relating to the semiconductor sector which has been a focus of our recent related work at Lincoln Network.
Increasing federal appropriations and incentives for semiconductor sector
The Commission urges Congress to establish and fund incentives and subsidies for the semiconductor sector, including funding the new programs authorized by the National Defense Authorization Act that passed in December. Specifically, it urged Congress to pass a 40 percent refundable investment tax credit for domestic semiconductor manufacturing. Further, the commissioners urged lawmakers to appropriate “at least $15 billion to subsidize several facilities in the United States to meet the end goal of multiple state-of-the-art sources for domestic fabrication.”
Aligning allies’ export controls and utilizing end-use controls to protect human rights
The Commission recommended an increased focus on export controls, including aligning international allies and updating our approach to export restrictions to protect human rights and ensure that American technology does not assist in the ongoing genocide. Specifically, it urged the State and Commerce Departments to partner with the Netherlands and Japan to align controls. Pointing to evidence that American-made chips were used for a supercomputer that’s used for mass surveillance in Xinjiang, the Commission recommended end-use controls to “prevent or deter U.S. firms from allowing certain key pieces of equipment, particularly high-end chips, to be utilized in malicious AI applications.”
Reforming and strengthening foreign investment security rules
The Commission also recommended that Congress review and update the law for reviewing foreign investments. In 2018, Congress passed the Foreign Investment Risk Review Modernization Act (FIRRMA), which updated the rules for the Committee on Foreign Investment in the United States. But the AI Commissioners warned that “CFIUS is not currently postured to address the range of threats that the United States faces from adversarial capital from strategic competitors such as China and Russia.” Specifically, it recommended establishing greater insight into Russia and Chinese investments in emerging technology, particularly for noncontrolling investments. The Commission urged additional CFIUS reforms and disclosures from investments made by “countries of special concern.” It also recommended a process for fast-tracking investments from trusted investors from allied countries.
Protecting the integrity of the U.S. research environment
The Commission smartly urged Congress and federal agencies to begin treating the U.S. research enterprise as a strategic national asset and afford it with similar protections that are made available to critical infrastructure owners and operators. It warned that China’s ongoing exploitation of American research institutions “violates the research community’s core principles of integrity, openness, accountability, and fairness,” and described federal response measures as “nascent.” The Commission recommended new rules for federal research grants and new information sharing tools, informed by the intelligence community, to help research institutions identify and prevent potential threats. They also called for a federally-sponsored, university-affiliated research center “to act as a center of excellence on research integrity and provide information and advice on research security.”
In addition, the Commission recommended additional cybersecurity support for research institutions, including the Department of Homeland Security’s Enhanced Cybersecurity Services program, which provides signature-based detection tools for critical infrastructure owners and operators. While that program has had a mixed track record, it is wise for the national security community to consider options for additional security measures to defend research institutions.
Improving visa vetting and promoting transparency about espionage threats
The Commission also urged the administration and national security community to take several actions to address potential counterintelligence threats before they reach research institutions and technology innovation hubs. For starters, it recommended stronger visa vetting and rules to restrict foreign nationals with ties to foreign military or intelligence services from gaining entry. The Commission also recommended transparency as a tool to deter this form of espionage, urging the creation of “an open-source database of organizations that have a history of improper technology transfer, intellectual property theft, or cyber espionage.” This would allow research institutions to do their own proactive vetting to improve security.
These are just a few of the highlights from the thorough and informative report. In the last Congress, the recommendations of the nonpartisan Cyberspace Solarium Commission were a catalyst for significant bipartisan reforms. Now, the 117th Congress and the Biden administration should look to the National Security Commission on AI’s valuable recommendations for the next generation of reforms to improve American security and promote our economic competitiveness.