|
on Artificial Intelligence |
| By: | Yiting Chen; Tracy Xiao Liu; You Shan; Songfa Zhong |
| Abstract: | As large language models (LLMs) like GPT become increasingly prevalent, it is essential that we assess their capabilities beyond language processing. This paper examines the economic rationality of GPT by instructing it to make budgetary decisions in four domains: risk, time, social, and food preferences. We measure economic rationality by assessing the consistency of GPT decisions with utility maximization in classic revealed preference theory. We find that GPT decisions are largely rational in each domain and demonstrate higher rationality scores than those of humans reported in the literature. We also find that the rationality scores are robust to the degree of randomness and demographic settings such as age and gender, but are sensitive to contexts based on the language frames of the choice situations. These results suggest the potential of LLMs to make good decisions and the need to further understand their capabilities, limitations, and underlying mechanisms. |
| Date: | 2023–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2305.12763&r=ain |
| By: | Samir Huseynov |
| Abstract: | This paper investigates the causal impact of negatively and positively framed ChatGPT Artificial Intelligence (AI) discussions on US students' anticipated labor market outcomes. Our findings reveal students reduce their confidence regarding their future earnings prospects after exposure to AI debates, and this effect is more pronounced after reading discussion excerpts with a negative tone. Unlike STEM majors, students in Non-STEM fields show asymmetric and pessimistic belief changes, suggesting that they might feel more vulnerable to emerging AI technologies. Pessimistic belief updates regarding future earnings are also prevalent across gender and GPA levels, indicating widespread AI concerns among all student subgroups. Educators, administrators, and policymakers may regularly engage with students to address their concerns and enhance educational curricula to better prepare them for a future that will be inevitably shaped by AI. |
| Date: | 2023–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2305.11900&r=ain |
| By: | Aman Saggu; Lennart Ante |
| Abstract: | The introduction of OpenAI's large language model, ChatGPT, catalyzed investor attention towards artificial intelligence (AI) technologies, including AI-related crypto assets not directly related to ChatGPT. Utilizing the synthetic difference-in-difference methodology, we identify significant 'ChatGPT effects' with returns of AI-related crypto assets experiencing average returns ranging between 10.7% and 15.6% (35.5% to 41.3%) in the one-month (two-month) period after the ChatGPT launch. Furthermore, Google search volumes, a proxy for attention to AI, emerged as critical pricing indicators for AI-related crypto post-launch. We conclude that investors perceived AI-assets as possessing heightened potential or value after the launch, resulting in higher market valuations. |
| Date: | 2023–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2305.12739&r=ain |
| By: | Sofia Patsali (Université Côte d'Azur, France; CNRS, GREDEG); Michele Pezzoni (Université Côte d'Azur, France; CNRS, GREDEG; Observatoire des Sciences et Techniques, HCERES, France; ICRIOS, Bocconi University, Italy); Jackie Krafft (Université Côte d'Azur, France; CNRS, GREDEG) |
| Abstract: | In line with the innovation procurement literature, this work investigates the impact of becoming a supplier of a national network of excellence regrouping French hospitals on the supplier's innovative performance. It investigates whether a higher information flow from hospitals to suppliers, proxied by the supply of AI-powered medical equipment, is associated with higher innovative performance. Our empirical analysis relies on a dataset combining unprecedented granular data on procurement bids and equipment with patent data to measure the firm's innovative performance. To identify the firm's innovative activities relevant to the bid, we use an advanced neural network algorithm for text analysis linking firms' equipment descriptions with relevant patent documents. Our results show that firms becoming hospital suppliers have a significantly higher propensity to innovate. About the mechanism, we show that supplying AI-powered equipment further boosts the suppliers' innovative performance, and this raises potential important policy implications. |
| Keywords: | Innovation performance, public procurement, medical equipment, hospitals, artificial intelligence |
| JEL: | H57 D22 O31 C81 |
| Date: | 2023–03 |
| URL: | http://d.repec.org/n?u=RePEc:gre:wpaper:2023-05&r=ain |
| By: | Brian Kan (Crean Lutheran High School, Irvine, United States); Douglas Klein (New Jersey City University, United States) |
| Abstract: | Amazon, with $469 Billion in sales in 2021, has established itself as a world-class user of AI, utilizing Machine Learning in its search engine to deliver desired results quickly - so millions of shoppers find the products they want to buy. Amazon’s affiliate, Amazon Web Services, had annual sales of $62 Billion in 2021, making it the 53rd largest company on the Fortune 500 as measured by revenues. AWS provides enterprises with a fully managed AI service with tools needed to execute every step of the ML development lifecycle in one integrated environment. By 2021, more than one hundred thousand companies utilized AWS Machine Learning - more than any other cloud platform. Outside of the traditional search engine applications what are some compelling and important business use cases where ML and AI have the greatest impact? Some use cases in this paper: AWS AI and Machine Learning are used by commercial landlords and industrial real estate owners to save energy and reduce carbon emissions. The World Wildlife Federation uses AWS AI tools in Indonesia to better understand the size and health of orangutan populations in their native habitat. And The Walt Disney Company uses ML and AI to organize metadata into one archival system, storing information about the stories, scenes, and characters in every second of Disney’s huge catalog of shows and movies. |
| Keywords: | AWS, Amazon Web Services AI, AWS Machine Learning, AWS Business Use Cases |
| Date: | 2022–06 |
| URL: | http://d.repec.org/n?u=RePEc:smo:raiswp:0205&r=ain |
| By: | Pawe{\l} Niszczota; Paul Conway |
| Abstract: | The introduction of ChatGPT has fuelled a public debate on the use of generative AI (large language models; LLMs), including its use by researchers. In the current work, we test whether delegating parts of the research process to LLMs leads people to distrust and devalue researchers and scientific output. Participants (N=402) considered a researcher who delegates elements of the research process to a PhD student or LLM, and rated (1) moral acceptability, (2) trust in the scientist to oversee future projects, and (3) the accuracy and quality of the output. People judged delegating to an LLM as less acceptable than delegating to a human (d = -0.78). Delegation to an LLM also decreased trust to oversee future research projects (d = -0.80), and people thought the results would be less accurate and of lower quality (d = -0.85). We discuss how this devaluation might transfer into the underreporting of generative AI use. |
| Date: | 2023–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2305.11873&r=ain |
| By: | Walter, Johannes |
| Abstract: | The EU's proposed Artificial Intelligence Act (AI Act) is meant to ensure safe AI systems in high-risk applications. The Act relies on human supervision of machine-learning algorithms, yet mounting evidence indicates that such oversight is not always reliable. In many cases, humans cannot accurately assess the quality of algorithmic recommendations, and thus fail to prevent harmful behaviour. This policy brief proposes three ways to solve the problem: First, Article 14 of the AI Act should be revised to acknowledge that humans often have difficulty assessing recommendations made by algorithms. Second, the suitability of human oversight for preventing harmful outcomes should be empirically tested for every high-risk application under consideration. Third, following Biermann et al. (2022), human decision-makers should receive feedback on past decisions to enable learning and improve future decisions. |
| Date: | 2023 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:zewpbs:022023&r=ain |
| By: | SOLER GARRIDO Josep (European Commission - JRC); FANO YELA Delia (European Commission - JRC); PANIGUTTI Cecilia (European Commission - JRC); JUNKLEWITZ Henrik (European Commission - JRC); HAMON Ronan (European Commission - JRC); EVAS Tatjana; ANDRÉ Antoine-Alexandre; SCALZO Salvatore |
| Abstract: | This report provides a systematic analysis of the current standardisation roadmap in support of the AI Act (AIA). The analysis covers standards currently considered by CEN-CENELEC Joint Technical Committee (JTC) 21 on artificial intelligence (AI), evaluating their coverage of the requirements laid out in the legal text. We found that the international standards currently considered already partially cover the AIA requirements for trustworthy AI defined in the regulation. Furthermore, many of the identified remaining gaps are already planned to be addressed by dedicated European standardisation. In order to support the work of standardisers in addressing these gaps, this document presents an independent expert-based analysis and recommendation, by highlighting areas deserving further attention of standardisers, and pointing, when possible, to additional relevant existing standards or directly providing possible additions to the scope of future European standards in support of the AI Act. |
| Keywords: | Artificial Intelligence, Standards, Transparency, Conformity Assessment, Risk Management, Data Quality, Human Oversight, Record Keeping, Quality Management, Robustness, Accuracy, Cybersecurity |
| Date: | 2023–05 |
| URL: | http://d.repec.org/n?u=RePEc:ipt:iptwpa:jrc132833&r=ain |
| By: | FARRELL Eimear (European Commission - JRC); GIUBILEI Miriam (European Commission - JRC); GRICIENE Asta; HARTOG Eddy; HUPONT TORRES Isabelle (European Commission - JRC); KOTSEV Alexander (European Commission - JRC); LOBO Georges; MARTINEZ RODRIGUEZ Eva (European Commission - JRC); SANDU Leontina; SCHADE Sven (European Commission - JRC); STROTMANN Maximilian; TANGI Luca (European Commission - JRC); TOLAN Songul; TORRECILLA SALINAS Carlos (European Commission - JRC); ULRICH Peter (European Commission - JRC) |
| Abstract: | The Public Sector plays different roles with regard to AI. First, it acts as regulator, establishing the legal framework for the use of AI within society. Second, governments play also the role of accelerator, providing funding and support for the uptake of AI. Third, public sector organisations develop and use Artificial Intelligence. To explore these roles, with particular emphasis on the latter, the Joint Research Centre (JRC) and the Directorate-General for Informatics (DIGIT) of the European Commission jointly organised a webinar series and a “science for policy” conference in 2022. This report includes the conclusions of each one of the webinars, together with the material and main findings of the closing event. It reveals recent challenges, opportunities, and policy perspectives of the use of AI in the public sector, and distils a set of short takeaway messages. In a nutshell these finding are (i) AI in the public sector implies multi-stakeholders; (ii) experiment first, scale-up later; (iii) trustworthiness is a must; (iv) there is a need for upskilling public sector to be ready for the AI revolution; and (v) adapt procurement for digital and AI innovation. The report concludes that the AI promise is high for the society and in particular for the Public Sector, but the risks are not to be minimized. Europe has the ambition to succeed as whole in the digital transition powered by data and by AI-based applications, and wants to do it the European way, by putting citizens in the centre of this transformation. |
| Keywords: | Artificial intelligence, digital transformation, innovative public services, Digital Decade |
| Date: | 2023–05 |
| URL: | http://d.repec.org/n?u=RePEc:ipt:iptwpa:jrc133826&r=ain |
| By: | Barry Quinn |
| Abstract: | This paper explores the journey of AI in finance, with a particular focus on the crucial role and potential of Explainable AI (XAI). We trace AI's evolution from early statistical methods to sophisticated machine learning, highlighting XAI's role in popular financial applications. The paper underscores the superior interpretability of methods like Shapley values compared to traditional linear regression in complex financial scenarios. It emphasizes the necessity of further XAI research, given forthcoming EU regulations. The paper demonstrates, through simulations, that XAI enhances trust in AI systems, fostering more responsible decision-making within finance. |
| Date: | 2023–06 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2306.02773&r=ain |
| By: | Vincent Lemaire; Gilles Pag\`es; Christian Yeo |
| Abstract: | We propose two parametric approaches to price swing contracts with firm constraints. Our objective is to create approximations for the optimal control, which represents the amounts of energy purchased throughout the contract. The first approach involves explicitly defining a parametric function to model the optimal control, and the parameters using stochastic gradient descent-based algorithms. The second approach builds on the first one, replacing the parameters with neural networks. Our numerical experiments demonstrate that by using Langevin-based algorithms, both parameterizations provide, in a short computation time, better prices compared to state-of-the-art methods (like the one given by Longstaff and Schwartz). |
| Date: | 2023–06 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2306.03822&r=ain |
| By: | Elias Zavitsanos; Dimitris Mavroeidis; Konstantinos Bougiatiotis; Eirini Spyropoulou; Lefteris Loukas; Georgios Paliouras |
| Abstract: | In this work, we examine the evaluation process for the task of detecting financial reports with a high risk of containing a misstatement. This task is often referred to, in the literature, as ``misstatement detection in financial reports''. We provide an extensive review of the related literature. We propose a new, realistic evaluation framework for the task which, unlike a large part of the previous work: (a) focuses on the misstatement class and its rarity, (b) considers the dimension of time when splitting data into training and test and (c) considers the fact that misstatements can take a long time to detect. Most importantly, we show that the evaluation process significantly affects system performance, and we analyze the performance of different models and feature types in the new realistic framework. |
| Date: | 2023–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2305.17457&r=ain |
| By: | Aadhitya A; Rajapriya R; Vineetha R S; Anurag M Bagde |
| Abstract: | Stock market is often important as it represents the ownership claims on businesses. Without sufficient stocks, a company cannot perform well in finance. Predicting a stock market performance of a company is nearly hard because every time the prices of a company stock keeps changing and not constant. So, its complex to determine the stock data. But if the previous performance of a company in stock market is known, then we can track the data and provide predictions to stockholders in order to wisely take decisions on handling the stocks to a company. To handle this, many machine learning models have been invented but they didn't succeed due to many reasons like absence of advanced libraries, inaccuracy of model when made to train with real time data and much more. So, to track the patterns and the features of data, a CNN-LSTM Neural Network can be made. Recently, CNN is now used in Natural Language Processing (NLP) based applications, so by identifying the features from stock data and converting them into tensors, we can obtain the features and then send it to LSTM neural network to find the patterns and thereby predicting the stock market for given period of time. The accuracy of the CNN-LSTM NN model is found to be high even when allowed to train on real-time stock market data. This paper describes about the features of the custom CNN-LSTM model, experiments we made with the model (like training with stock market datasets, performance comparison with other models) and the end product we obtained at final stage. |
| Date: | 2023–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2305.14378&r=ain |
| By: | Simerjot Kaur; Andrea Stefanucci; Sameena Shah |
| Abstract: | Determining industry and product/service codes for a company is an important real-world task and is typically very expensive as it involves manual curation of data about the companies. Building an AI agent that can predict these codes automatically can significantly help reduce costs, and eliminate human biases and errors. However, unavailability of labeled datasets as well as the need for high precision results within the financial domain makes this a challenging problem. In this work, we propose a hierarchical multi-class industry code classifier with a targeted multi-label product/service code classifier leveraging advances in unsupervised representation learning techniques. We demonstrate how a high quality industry and product/service code classification system can be built using extremely limited labeled dataset. We evaluate our approach on a dataset of more than 20, 000 companies and achieved a classification accuracy of more than 92\%. Additionally, we also compared our approach with a dataset of 350 manually labeled product/service codes provided by Subject Matter Experts (SMEs) and obtained an accuracy of more than 96\% resulting in real-life adoption within the financial domain. |
| Date: | 2023–05 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2305.13532&r=ain |
| By: | Julia M. Puaschunder (The New School, New York, USA) |
| Abstract: | The Artificial Intelligence (AI) evolution is a broad set of methods, algorithms, and technologies making software human-like intelligent that is encroaching our contemporary workplace. Thinking like humans but acting rational is the primary goal of AI innovations. The current market disruption with AI lies at the core of the IT-enhanced economic growth driven by algorithms – for instance enabled via the sharing economies and big data information gains, self-check outs, online purchases and bookings, medical services social care, law, retail, logistics and finance to name a few domains in which AI leads to productivity enhancement. While we have ample account of AI entering our everyday lives, we hardly have any information about economic growth driven by AI. Preliminary studies found a negative relation between digitalization and economic growth, indicating that we lack a proper growth theory capturing the economic value imbued in AI. We also have information that indicates AI-led growth based on ICT technologies may widen an inequality-rising skilled versus unskilled labor wage gap. This paper makes the theoretical case of AI as a self-learning entity to be integrated into endogenous growth theory, which gives credit to learning and knowledge transformation as vital economic productivity ingredients. Future research may empirically validate the claim that AI as a self-learning entity is a driver of endogenous growth. All these endeavors may prepare for research on how to enhance human welfare with AI-induced growth based on inclusive AI-human compatibility and mutual exchange between machines and human beings. |
| Keywords: | Algorithms, Artificial Intelligence (AI), Digitalization, Digitalization disruption, Digital inequality, Economic growth, Endogenous growth |
| Date: | 2022–10 |
| URL: | http://d.repec.org/n?u=RePEc:smo:raiswp:0224&r=ain |
| By: | Julia M. Puaschunder (Columbia University, Graduate School of Arts and Sciences) |
| Abstract: | We live in the age of digitalization. Digital disruption is the advancement of our lifetimes. Never before in the history of humankind have human beings given up as much decision-making autonomy as today to a growing body of artificial intelligence (AI). Digitalization features a wave of self-learning entities that generate information from exponentially-growing big data sources that are encroaching every aspect of our daily lives. Inequality is one of the most significant pressing concern of our times. Ample evidence exists in economics, law and historical studies that multiple levels of inequality dominate the current socio-dynamics, politics and living conditions around the world. Social inequality stretches from societal levels within nation states to global dimensions but also intergenerational inequality domains. While digitalization and inequality are predominant features of our times, hardly any information exists on the inequality inherent in digitalization. This paper breaks new ground in theoretically arguing for inequality being an overlooked by-product of innovative change – featuring concrete examples in insights and applications in the digitalization domain. A multi-faceted analysis will draw a contemporary digital inequality account from behavioral economic, macroeconomic, comparative and legal economic perspectives. This paper targets at aiding academics and practitioners in understanding the advantages but also the potential inequalities imbued in digitalization. It sets a historic landmark to capture the Zeitgeist of our digitalization disruption heralding unexpected inequalities stemming from innovative change. The article may open eyes to understand our times holistically in its advantageous innovation capacities but also potential societal, international and intertemporal unequal gains and losses perspectives from digitalization. |
| Keywords: | AI, Artificial Intelligence, Behavioral Economics, Behavioral Macroeconomics, Big Data, Big Data Insights, Coronavirus crisis |
| Date: | 2022–06 |
| URL: | http://d.repec.org/n?u=RePEc:smo:raiswp:0201&r=ain |
| By: | Julia M. Puaschunder (Columbia University, Graduate School of Arts and Sciences, USA) |
| Abstract: | Intergenerational Equity has become the basis for legal codification in different domains in the most recent decades. This article discusses intergenerational representations in the judiciary system and democracy. Age-balanced decision-making of juries that could transpire into the judiciary system, democratic rotation principles with age-sensitive quotas but also futuristic Artificial Intelligence-led governance that pays attention to the mean age of voting circles are prospective intergenerational justice and democracy advancements proposed. |
| Keywords: | AI, Artificial Intelligence, Behavioral Economics, Behavioral Macroeconomics, Democracy, Digitalization, Discounting, Disparate Impact, Economics, Equality, Intergenerational Democracy, |
| Date: | 2022–06 |
| URL: | http://d.repec.org/n?u=RePEc:smo:raiswp:0190&r=ain |