​​This Classifying AI Systems Interactive presents several AI system classification frameworks developed to distill AI systems into concise, comparable and policy-relevant dimensions. It provides key takeaways and framework-specific results from CSET’s analysis of more than 1,800 system classifications done by survey respondents using the frameworks. You can explore the frameworks and example AI systems used in the survey, and even take the survey.

This paper is the fourth installment in a series on “AI safety,” an area of machine learning research that aims to identify causes of unintended behavior in machine learning systems and develop tools to ensure these systems work safely and reliably. The first paper in the series, “Key Concepts in AI Safety: An Overview,” outlined three categories of AI safety issues—problems of robustness, assurance, and specification—and the subsequent two papers described problems of robustness and assurance, respectively. This paper introduces specification as a key element in designing modern machine learning systems that operate as intended.

This data brief explores how international collaboration relates to the impact and output of research publications. Focusing on the top 10 countries with the highest publication output from 2010 to 2019, the authors provide a comprehensive analysis across the major fields of science and technology.

This brief explores the development and testing of artificial intelligence system classification frameworks intended to distill AI systems into concise, comparable and policy-relevant dimensions. Comparing more than 1,800 system classifications, it points to several factors that increase the utility of a framework for human classification of AI systems and enable AI system management, risk assessment and governance.

Militaries around the world have often relied on the largest global defense companies to acquire and integrate cutting-edge technologies. This issue brief examines the investment and mergers and acquisition activities in artificial intelligence of the top 50 global defense companies — a key, if limited, approach to accessing AI innovation in the commercial sector — and assesses investment trends of their corporate venture capital subsidiaries and offers a geographic breakdown of defense companies and their AI target companies.

As U.S. policymakers grapple with the need to control international technology flows, this annotated bibliography distills key lessons and surveys 50 years of scholarship, government documents, and commentary. The resources it presents are at the intersection of international economics and technology and span from the Cold War to the current challenges surrounding U.S.-China relations.

Conventional wisdom suggests that cutting-edge artificial intelligence is dependent on large volumes of data. An overemphasis on “big data” ignores the existence—and underestimates the potential—of several AI approaches that do not require massive labeled datasets. This issue brief is a primer on “small data” approaches to AI. It presents exploratory findings on the current and projected progress in scientific research across these approaches, which country leads, and the major sources of funding for this research.

As dual-use technologies transform the national security landscape, the U.S. Department of Defense has established a variety of offices and programs dedicated to bringing private sector innovation into the military. However, these efforts have largely failed to drive cutting-edge commercial technology into major military platforms and systems. This report examines the shortcomings of the DOD’s current approach to defense innovation and offers recommendations for a more effective strategy.

For the first time, a Chinese company—Huawei—is set to lead the global transition from one key national security infrastructure technology to the next. How did Washington, at the beginning of the twenty-first century, fail to protect U.S. firms in this strategic technology and allow a geopolitical competitor to take a leadership position in a national security relevant critical infrastructure such as telecommunications? This policy brief highlights the characteristics of 5G development that China leveraged, exploited, and supported to take the lead in this key technology. The Huawei case study is in some ways the canary in the coal mine for emerging technologies and an illustration of what can happen to U.S. competitiveness when China’s companies do not have to base decisions on market forces.

As modern machine learning systems become more widely used, the potential costs of malfunctions grow. This policy brief describes how trends we already see today—both in newly deployed artificial intelligence systems and in older technologies—show how damaging the AI accidents of the future could be. It describes a wide range of hypothetical but realistic scenarios to illustrate the risks of AI accidents and offers concrete policy suggestions to reduce these risks.