July 26, 2023

Sharper: AI Safety and Stability 

Analysis from CNAS experts on the most critical challenges for U.S. foreign policy.

Artificial intelligence has sprawled its way across national news headlines, leading to a range of pressing conversations about the technology and how U.S. policymakers should get ahead of its wide-ranging applications. Should there be an international watchdog for AI? How should the United States regulate cutting-edge AI models? Can America win the AI race? CNAS experts are sharpening the conversation around AI’s role in national security. Continue reading this edition of Sharper to explore their ideas and recommendations.

Features

U.S.-China Competition and Military AI

Two tectonic trends in the international security environment appear to be on a collision course. The first trend is the intensifying geopolitical rivalry between the United States and the People’s Republic of China. The second trend is the rapid development of artificial intelligence (AI) technologies, including for military applications. A new report from authors Jacob Stokes and Alexander Sullivan, with Noah Greene explores the intersection of these trends through five pathways through which military AI could potentially undermine stability and increase strategic risks between the United States and China.

Response to OSTP “National Priorities for Artificial Intelligence Request for Information”

In May 2023, the White House Office of Science and Technology Policy (OSTP) issued a Request for Information (RFI) seeking input on national priorities on AI. CNAS experts submitted the following response, with a particular focus on policy measures for ‘frontier’ AI systems—general-purpose ‘foundation’ models at the frontier of capabilities and risks. To manage national security risks from these models, the response outlines recommendations across the AI lifecycle—from supporting technical AI safety research, through to risk assessments and evaluations around model release—and around the role of regulatory bodies and federal government policy levers.

Indo-Pacific Security

U.S.-China Competition and Military AI

Two tectonic trends in the international security environment appear to be on a collision course. The first trend is the intensifying geopolitical rivalry between the United S...

Technology & National Security

Response to OSTP “National Priorities for Artificial Intelligence Request for Information”

In May 2023, the White House Office of Science and Technology Policy issued a Request for Information seeking input on national priorities on AI. CNAS experts submitted the fo...

Frontier AI Regulation: Managing Emerging Risks to Public Safety

"Guardrails are necessary to prevent the pursuit of innovation from imposing excessive negative externalities on society," write Tim Fist, Markus Anderljung, and other authors in an new report published on ArXiv. "There is increasing recognition that government oversight is needed to ensure AI development is carried out responsibly; we hope to contribute to this conversation by exploring regulatory approaches to this end."

Biden-⁠Harris Administration Secures Voluntary Commitments from Leading Artificial Intelligence Companies to Manage the Risks Posed by AI

In this edition of Noteworthy, researchers from the AI Safety and Stability team comment on the July 2023 announcement from the White House on securing voluntary commitments on safety, security, and trust from leading AI companies.

Technology & National Security

Frontier AI Regulation: Managing Emerging Risks to Public Safety

We think that it is important to begin taking practical steps to regulate frontier AI today, and that the ideas discussed in this paper are a step in that direction....

Technology & National Security

NOTEWORTHY: Biden-⁠Harris Administration Secures Voluntary Commitments from Leading Artificial Intelligence Companies to Manage the Risks Posed by AI

In this edition of Noteworthy, researchers from the AI Safety and Stability team comment on the announcement from the White House on securing voluntary commitments on safety, ...

Commentaries

AI’s Gatekeepers Aren’t Prepared for What’s Coming

"Until recently, AI has been a diffuse technology that rapidly proliferates," writes Paul Scharre in Foreign Policy. "Open-source AI models are readily available online. The recent shift to large models, such as OpenAI’s ChatGPT, is concentrating power in the hands of large tech companies that can afford the computing hardware needed to train these systems. The balance of global AI power will hinge on whether AI concentrates power in the hands of a few actors, as nuclear weapons did, or proliferates widely, as smartphones have."

How to Win Friends and Choke China’s Chip Supply

"While the United States has been careful to justify the new strategy on national security grounds, the strategy will unquestionably have significant economic impacts on China, which will compound over time as China’s capabilities are frozen in place and lag further and further behind other global producers," argues Emily Kilcrease in War on the Rocks. "It also remains unclear how far the U.S. strategy will extend beyond chips. The administration has identified artificial intelligence, quantum information systems, biotechnology and biomanufacturing, and advanced clean energy technologies as fundamental to U.S. national security. If the United States imposed restrictions across these technology ecosystems comparable to what has just been done for advanced chips production and supercomputing, it would almost certainly lead to a broad technological decoupling from China."

Technology & National Security

AI’s Gatekeepers Aren’t Prepared for What’s Coming

What is clear is that current state-of-the-art AI models are not safe and no one knows how to reliably make them safe....

Energy, Economics & Security

How to Win Friends and Choke China’s Chip Supply

U.S. officials should prioritize aligning controls with other major producer nations to target the most significant technology chokepoints and the areas where the risk of back...

Chinese Firms Are Evading Chip Controls

"Enforcing a ban on Chinese access to leading-edge chips won’t be easy," observe Tim Fist, Jordan Schneider and Lennart Heim in Foreign Policy. "But by establishing a chip inspection program, taking steps to help ensure cloud services are secure, providing the BIS with more funding, and continuing to seek allied cooperation, the U.S. government can erect significant barriers to China using advanced computing to power a new generation of dangerous military applications."

China Is Flirting with AI Catastrophe

"As the world settles into a new era of rivalry­—this time between China and the United States—competition over another revolutionary technology, artificial intelligence, has sparked a flurry of military and ethical concerns parallel to those initiated by the nuclear race," argue Bill Drexel and Hannah Kelley in Foreign Affairs. "Those concerns are well worth the attention they are receiving, and more: a world of autonomous weapons and machine-speed war could have devastating consequences for humanity. Beijing’s use of AI tools to help fuel its crimes against humanity against the Uyghur people in Xinjiang already amounts to a catastrophe."

Technology & National Security

Chinese Firms Are Evading Chip Controls

The U.S. government can erect significant barriers to China using advanced computing to power a new generation of dangerous military applications....

Technology & National Security

China Is Flirting With AI Catastrophe

Today’s AI sprint would not be the first time Beijing’s desire to hasten progress invited disaster....

AI Nuclear Weapons Catastrophe Can Be Avoided

"This scenario has ample room for catastrophe—for both the international community and domestic leaders—should any of these weapons malfunction and lead to a nuclear fallout," warns Noah Greene in Lawfare. "As the Soviet-era Col. Petrov case kindly taught us, without a human firmly in control of the nuclear command-and-control structure, the odds of disaster creep slowly toward an unintended or uncontrolled nuclear exchange. An agreement between nuclear powers on this issue led by P5 states would be an important step toward recreating a patchwork of nuclear treaties that has dissolved over the past two decades. To do otherwise would be to flirt with an AI-enabled nuclear arms race."

An AI Challenge: Balancing Open and Closed Systems

"The open vs. closed framework remains imperfect," observes Pablo Chavez in CEPA. "In the AI context, both sides have a legitimate claim to being a better model for safety and security. As noted by Stability AI, an open-source AI company, open models and datasets can help ensure robust oversight; third parties with access can anticipate emerging risks and implement mitigations. But the nature of open-source licensing (which generally makes source code available to all) opens the door to entities or individuals who wish to cause harm or are simply not concerned about or resourced for risk mitigation."

Roles and Implications of AI in the Russian-Ukrainian Conflict

"Artificial Intelligence (AI) is emerging as a significant asset in the ongoing Russian-Ukrainian conflict," writes Samuel Bendett in Russia Matters. "Specifically, it has become a key data analysis tool that helps operators and warfighters make sense of the growing volume and amount of information generated by numerous systems, weapons and soldiers in the field. As AI use continues to evolve, its application on the current Ukrainian and future battlefields will translate into more precise and capable responses to adversary forces, movements and actions."

Technology & National Security

AI Nuclear Weapons Catastrophe Can Be Avoided

AI-enabled nuclear weapons are particularly concerning due to their civilization-destroying nature....

Technology & National Security

An AI Challenge: Balancing Open and Closed Systems

Navigating the spectrum between open and closed is critical to effective artificial intelligence policy. The right balance will promote innovation and competition while managi...

Technology & National Security

Roles and Implications of AI in the Russian-Ukrainian Conflict

Artificial Intelligence (AI) is emerging as a significant asset in the ongoing Russian-Ukrainian conflict. Specifically, it has become a key data analysis tool that helps oper...

In the News:

Featuring commentary and analysis from experts including Paul Scharre, Richard Fontaine, Tim Fist, and Lt. Gen. Jack Shanahan.

Technology & National Security

How Militaries Are Using Artificial Intelligence on and off the Battlefield

Artificial intelligence has been a crucial tool for many nations’ militaries for years. Now, with the war in Ukraine driving innovation, AI’s role is likely to grow. Paul Scha...

Technology & National Security

Ai Threat Landscape Could Include Automated Propaganda Bots, Sophisticated Email Attacks: Security Experts

Center for a New American Security CEO Richard Fontaine told Fox News Digital that until now, humans have primarily created disinformation. While it may have been propagated t...

Technology & National Security

U.S. Grapples With Potential Threats From Chinese AI

Any policy banning investment such as Qualcomm’s would be challenging to articulate and enforce. Simply examining the code of a given AI system wouldn’t necessarily reveal its...

Technology & National Security

Pentagon should experiment with AIs like ChatGPT — but don’t trust them yet: DoD’s ex-AI chiefs

“I’m excited about the potential,” said Lt. Gen. (ret.) Jack Shanahan, founding director of the Pentagon’s Joint Artificial Intelligence Center (JAIC) from 2018 to 2020. “I pl...

About the Sharper Series

The CNAS Sharper series features curated analysis and commentary from CNAS experts on the most critical challenges in U.S. foreign policy. From the future of America's relationship with China to the state of U.S. sanctions policy and more, each collection draws on the reports, interviews, and other commentaries produced by experts across the Center to explore how America can strengthen its competitive edge.

Subscribe

Sign up to receive the latest analysis from the CNAS expert community on the most important issues facing America's national security.





  • Reports
    • June 11, 2024
    Catalyzing Crisis

    Executive Summary The arrival of ChatGPT in November 2022 initiated both great excitement and fear around the world about the potential and risks of artificial intelligence (A...

    By Bill Drexel & Caleb Withers

  • Commentary
    • Just Security
    • June 6, 2024
    Open Source AI: The Overlooked National Security Imperative

    Now a global technological superpower, China does not want to repeat the mistakes of its past and is actively positioning itself to be the world’s AI leader....

    By Keegan McBride

  • Commentary
    • The Washington Post
    • May 30, 2024
    To Win the Chip War, the U.S. Must Prioritize Revolutionary Research

    Taking big bets on moonshot technologies is the only approach that can sustain Moore’s law and guarantee that the United States continues to lead in the technologies of tomorr...

    By Jordan Schneider, Arrian Ebrahimi & Chris Miller

  • Reports
    • May 28, 2024
    The Quest for Qubits

    Executive Summary The United Nations General Assembly recently declared 2025 the International Year of Quantum Science and Technology.1 Quantum information science, a field on...

    By Sam Howell

View All Reports View All Articles & Multimedia