While recent progress in artificial intelligence (AI) has relied primarily on increasing the size and scale of the models and computing budgets for training, we ask if those trends will continue. Financial incentives are against scaling, and there can be diminishing returns to further investment. These effects may already be slowing growth among the very largest models. Future progress in AI may rely more on ideas for shrinking models and inventive use of existing models than on simply increasing investment in compute resources.

Concerns over risks from generative artificial intelligence systems have increased significantly over the past year, driven in large part by the advent of increasingly capable large language models. But, how do AI developers attempt to control the outputs of these models? This primer outlines four commonly used techniques and explains why this objective is so challenging.

Recent government directives, international conferences, and media headlines reflect growing concern that artificial intelligence could exacerbate biological threats. When it comes to biorisk, AI tools are cited as enablers that lower information barriers, enhance novel biothreat design, or otherwise increase a malicious actor’s capabilities. In this explainer, CSET Biorisk Research Fellow Steph Batalis summarizes the state of the biorisk landscape with and without AI.

CSET submitted the following comment in response to a Request for Comment (RFC) from the Office of Management and Budget (OMB) about a draft memorandum providing guidance to government agencies regarding the appointment of Chief AI Officers, Risk Management for AI, and other processes following the October 30, 2023 Executive Order on AI.

Demand for talent is one of the core elements of technological competition between the United States and China. In this issue brief, we explore demand signals in China’s domestic AI workforce in two ways: geographically and within the defense and surveillance sectors. Our exploration of job postings from Spring 2021 finds that more than three-quarters of all AI job postings are concentrated in just three regions: the Yangtze River Delta region, the Pearl River Delta, and the Beijing-Tianjin-Hebei area.

Antimicrobial resistance (AMR) is one of the world’s most pressing global health threats. Basic research is the first step towards identifying solutions. This brief examines the AMR research landscape since 2000, finding that the amount of research is increasing and that the U.S. is a leading publisher, but also that novel solutions like phages and synthetic antimicrobial production are a small portion of that research.

This report summarizes recent DOD digital workforce trends as an update to our 2021 report. We expanded our definition of AI talent to include data, analytics, software, and AI, referred to here as the “digital workforce,” to be more aligned with Department needs and current workforce planning efforts. We find that the Department of Defense continues to struggle with their ability to define, identify, develop, assign, promote, and retain digital talent.

AI capabilities are evolving quickly and pose novel—and likely significant—risks. In these rapidly changing conditions, how can policymakers effectively anticipate and manage risks from the most advanced and capable AI systems at the frontier of the field? This Roundtable Report summarizes some of the key themes and conclusions of a July 2023 workshop on this topic jointly hosted by CSET and Google DeepMind.

CSET submitted the following comment in response to a Request for Information (RFI) from the White House's Office of Science and Technology Policy about potential changes to the Policies for Federal and Institutional Oversight of Life Sciences Dual Use Research of Concern (DURC) and Recommended Policy Guidance for Departmental Development of Review Mechanisms for Potential Pandemic Pathogen Care and Oversight (P3CO).

This explainer overviews techniques to produce smaller and more efficient language models that require fewer resources to develop and operate. Importantly, information on how to leverage these techniques, and many of the subsequent small models, are openly available online for anyone to use. The combination of both small (i.e., easy to use) and open (i.e., easy to access) could have significant implications for artificial intelligence development.