|
on Artificial Intelligence |
| By: | Benjamin S. Manning; Kehang Zhu; John J. Horton |
| Abstract: | We present an approach for automatically generating and testing, in silico, social scientific hypotheses. This automation is made possible by recent advances in large language models (LLM), but the key feature of the approach is the use of structural causal models. Structural causal models provide a language to state hypotheses, a blueprint for constructing LLM-based agents, an experimental design, and a plan for data analysis. The fitted structural causal model becomes an object available for prediction or the planning of follow-on experiments. We demonstrate the approach with several scenarios: a negotiation, a bail hearing, a job interview, and an auction. In each case, causal relationships are both proposed and tested by the system, finding evidence for some and not others. We provide evidence that the insights from these simulations of social interactions are not available to the LLM purely through direct elicitation. When given its proposed structural causal model for each scenario, the LLM is good at predicting the signs of estimated effects, but it cannot reliably predict the magnitudes of those estimates. In the auction experiment, the in silico simulation results closely match the predictions of auction theory, but elicited predictions of the clearing prices from the LLM are inaccurate. However, the LLM's predictions are dramatically improved if the model can condition on the fitted structural causal model. In short, the LLM knows more than it can (immediately) tell. |
| Date: | 2024–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2404.11794&r=ain |
| By: | Gasteiger, Emanuel; Kuhn, Michael; Mistlbacher, Matthias; Prettner, Klaus |
| Abstract: | While automation technologies replace workers in ever more tasks, robots, 3D printers, and AI-based applications require substantial amounts of electricity. This raises concerns regarding the feasibility of the energy transition towards mitigating climate change. How does automation interact with conventional capital in driving energy demand and how do taxes on robots and taxes on electricity affect the adoption of robots and AI? To answer these questions, we generalize a standard economic growth model with automation and electricity use. In addition, we augment the model with electricity taxes and robot taxes and show the mechanisms by which these taxes affect automation. We find that an electricity tax serves a similar purpose as a robot tax. However, a robot tax is much more difficult to implement from a practical perspective. |
| Keywords: | Automation; Robots; Growth; Electricity Use; Energy Taxes; Robot Taxes |
| Date: | 2024–04 |
| URL: | http://d.repec.org/n?u=RePEc:wiw:wus005:62095883&r=ain |
| By: | R. Michael Alvarez; Jacob Morrier |
| Abstract: | This paper presents a new approach to evaluating the quality of answers in political question-and-answer sessions. We propose to measure an answer's quality based on the degree to which it allows us to infer the initial question accurately. This conception of answer quality inherently reflects their relevance to initial questions. Drawing parallels with semantic search, we argue that this measurement approach can be operationalized by fine-tuning a large language model on the observed corpus of questions and answers without additional labeled data. We showcase our measurement approach within the context of the Question Period in the Canadian House of Commons. Our approach yields valuable insights into the correlates of the quality of answers in the Question Period. We find that answer quality varies significantly based on the party affiliation of the members of Parliament asking the questions and uncover a meaningful correlation between answer quality and the topics of the questions. |
| Date: | 2024–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2404.08816&r=ain |
| By: | Yi Chen; Hanming Fang; Yi Zhao; Zibo Zhao |
| Abstract: | Categorical variables have no intrinsic ordering, and researchers often adopt a fixed-effect (FE) approach in empirical analysis. However, this approach has two significant limitations: it overlooks textual labels associated with the categorical variables; and it produces unstable results when there are only limited observations in a category. In this paper, we propose a novel method that utilizes recent advances in large language models (LLMs) to recover overlooked information in categorical variables. We apply this method to investigate labor market mismatch. Specifically, we task LLMs with simulating the role of a human resources specialist to assess the suitability of an applicant with specific characteristics for a given job. Our main findings can be summarized in three parts. First, using comprehensive administrative data from an online job posting platform, we show that our new match quality measure is positively correlated with several traditional measures in the literature, and at the same time, we highlight the LLM's capability to provide additional information conditional on the traditional measures. Second, we demonstrate the broad applicability of the new method with a survey data containing significantly less information than the administrative data, which makes it impossible to compute most of the traditional match quality measures. Our LLM measure successfully replicates most of the salient patterns observed in a hard-to-access administrative dataset using easily accessible survey data. Third, we investigate the gender gap in match quality and explore whether there exists gender stereotypes in the hiring process. We simulate an audit study, examining whether revealing gender information to LLMs influences their assessment. We show that when gender information is disclosed to the GPT, the model deems females better suited for traditionally female-dominated roles. |
| JEL: | C55 J16 J24 J31 |
| Date: | 2024–04 |
| URL: | http://d.repec.org/n?u=RePEc:nbr:nberwo:32327&r=ain |
| By: | Böcking, Lars; Michaelis, Anne; Schäfermeier, Bastian; Baier, André; Kühl, Niklas; Körner, Marc-Fabian; Nolting, Lars |
| Keywords: | Generative Künstliche Intelligenz, GenAI, Energiewirtschaft, TenneT |
| Date: | 2024 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:bayism:290410&r=ain |
| By: | Alessia Lo Turco (Department of Economics and Social Sciences, Universita' Politecnica delle Marche (UNIVPM)); Alessandro Sterlacchini (Department of Economics and Social Sciences, Universita' Politecnica delle Marche) |
| Abstract: | In this paper we consider firms involved in two waves (2019 and 2021) of the French ICT survey to distinguish between early and late adopters of AI technologies and to highlight some relevant antecedents that facilitated the former to keep and the latter to start adopting them. The implementation of data security systems, the training and recruitment of employees for ICT, and the use of websites and social media for collecting information on customers, increase the probability of keeping and starting the AI adoption. We also show that the impact of these factors differs according to the business function AI technologies are used for. They appear to be more relevant for the administration and marketing functions. Furthermore, the usage of AI for marketing is also fostered by the antecedent use of e-commerce and CRM applications. These findings support the hypothesis that the AI adoption by firms is shaped by a hierarchical trajectory, from less to more complex and demanding technologies in terms of complementary investments in ICT and skills. |
| Keywords: | Artificial Intelligence, Digital technologies and skills, IT security systems, French firms. |
| JEL: | O31 O33 |
| Date: | 2024–04 |
| URL: | http://d.repec.org/n?u=RePEc:anc:wpaper:486&r=ain |
| By: | Alicia Vidler |
| Abstract: | Traditionally, assets are selected for inclusion in a portfolio (long or short) by human analysts. Teams of human portfolio managers (PMs) seek to weigh and balance these securities using optimisation methods and other portfolio construction processes. Often, human PMs consider human analyst recommendations against the backdrop of the analyst's recommendation track record and the applicability of the analyst to the recommendation they provide. Many firms regularly ask analysts to provide a "conviction" level on their recommendations. In the eyes of PMs, understanding a human analyst's track record has typically come down to basic spread sheet tabulation or, at best, a "virtual portfolio" paper trading book to keep track of results of recommendations. Analysts' conviction around their recommendations and their "paper trading" track record are two crucial workflow components between analysts and portfolio construction. Many human PMs may not even appreciate that they factor these data points into their decision-making logic. This chapter explores how Artificial Intelligence (AI) can be used to replicate these two steps and bridge the gap between AI data analytics and AI-based portfolio construction methods. This field of AI is referred to as Recommender Systems (RS). This chapter will further explore what metadata that RS systems functionally supply to downstream systems and their features. |
| Date: | 2024–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2404.11080&r=ain |
| By: | Masanori Hirano; Kentaro Imajo |
| Abstract: | Large language models (LLMs) are now widely used in various fields, including finance. However, Japanese financial-specific LLMs have not been proposed yet. Hence, this study aims to construct a Japanese financial-specific LLM through continual pre-training. Before tuning, we constructed Japanese financial-focused datasets for continual pre-training. As a base model, we employed a Japanese LLM that achieved state-of-the-art performance on Japanese financial benchmarks among the 10-billion-class parameter models. After continual pre-training using the datasets and the base model, the tuned model performed better than the original model on the Japanese financial benchmarks. Moreover, the outputs comparison results reveal that the tuned model's outputs tend to be better than the original model's outputs in terms of the quality and length of the answers. These findings indicate that domain-specific continual pre-training is also effective for LLMs. The tuned model is publicly available on Hugging Face. |
| Date: | 2024–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2404.10555&r=ain |
| By: | Shiro ARMSTRONG; Jacob TAYLOR |
| Abstract: | The digital economy and artificial intelligence (AI) play an increasingly pivotal role in global economic and societal value creation but lack multilateral rules. The current fragmented state of the global digital economy risks dampening digital and AI systems’ productivity growth potential and exacerbating their emerging risks. This paper provides three building blocks for new approaches to multilateral governance that can more evenly distribute the benefits of digital and AI systems while collectively managing their risks. First, it analyzes the economic logic of value creation in the digital economy and the policy dilemmas that this logic implies. Second, it identifies major economic and political challenges that impede efforts to advance multilateral governance, including concentration of power, protectionism, and exclusion in digital and AI systems. Third, it evaluates the potential of Digital Public Infrastructure (DPI)—an increasingly globally-recognized framework for promoting publicly guaranteed digital ecosystems—to serve as a foundation for more equitable, interoperable and inclusive global digital and AI governance. The paper concludes by identifying near-term opportunities for policymakers to align on shared multilateral principles while respecting all countries' domestic policy space. |
| Date: | 2024–04 |
| URL: | http://d.repec.org/n?u=RePEc:eti:dpaper:24052&r=ain |