The Digital Services Act, adopted in 2022, updated the liability rules for online intermediaries and introduced obligations for Very Large Online Platforms (VLOPs) and Very Large Online Search Engines (VLOSEs) to conduct systemic risk assessments in four broad areas of risk to society. While systemic risk assessments have long been carried out in other sectors such as financial services, banking, or insurance, the introduction of this type of assessments in the context of digital services is a novelty. In this new CERRE Tech, Media, & Telecom project, authors Sally Broughton Micova and Andrea Calef draw on the literature on systemic risk and its application in financial services and banking to contribute to the understanding and implementation of systemic risk assessments under the DSA. How has the concept of systemic risk been defined and used in financial services? What can be learnt from this to understand how VLOPs and VLOSEs may pose a risk of harm?
28 May 2026 in Proceedings of the National Academy of Sciences
Large Language Models (LLMs) such as ChatGPT are transforming how scientists conduct and validate research, offering promise as tools to improve scientific reproducibility. However, computational reproducibility and error detection remain expensive and labor-intensive. We experimentally test how collaboration between researchers and LLM assistants influences the reproduction of quantitative social science findings across different levels of AI autonomy. We randomly assigned 288 researchers to 103 teams working under three conditions: human-only, AI-assisted (using ChatGPT as a collaborative tool), or AI-led (ChatGPT operating with minimal human oversight). Teams reproduced published results from leading social science journals, detected coding errors, and proposed robustness checks. Human-only and AI-assisted teams achieved comparable reproduction rates (94% vs. 91%) and performed similarly on most outcomes, except human-only teams identified significantly more major coding errors. Both substantially outperformed AI-led teams, which achieved only a 37% reproduction rate, detected fewer errors across all categories, proposed weaker robustness checks, and required more time. This autonomous approach, however, likely represents only a lower bound of AI capabilities. Despite rapid model advances, expert human judgment currently remains indispensable for reliable empirical verification. While AI assistance did not degrade most outcomes, it provided no measurable advantages and was associated with reduced detection of major errors. However, the 37% autonomous reproduction rate indicates that AI could provide value in settings where scale or cost constraints preclude human review of papers, even though general-purpose LLMs offer no immediate advantages for human-supervised verification.
In this paper, we investigate whether and to what extent credit market freedom affects a firm’s target level of investment, cash holdings, and leverage. To do so, we generalize the standard empirical models, commonly used in the finance literature to estimate those targets, in order to incorporate credit market freedom into the set of regressors. We estimate three augmented models on a large and heterogeneous sample of North American nonfinancial firms over the period 2000–2019. Our empirical results suggest that greater credit market freedom is associated with a healthier corporate capital structure, higher financial flexibility, and a friendlier investment environment. Our paper contributes to both economic freedom and finance literatures by investigating an unexplored issue in economics and corporate finance research. In addition, it informs policymakers that promoting financial reforms that increase credit market freedom can boost a country’s economic growth.