Paper by Zhen-Song Chen and Zheng Ma: “Ensuring fair and accurate hiring outcomes is critical for both job seekers’ economic opportunities and organizational development. This study addresses the challenge of mitigating biases in AI-powered resume screening systems by leveraging crowd intelligence, thereby enhancing problem-solving efficiency and decision-making quality. We propose a novel counterfactual resume-annotation method based on a causal model to capture and correct biases from human resource (HR) representatives, providing robust ground truth data for supervised machine learning. The proposed model integrates multiple language embedding models and diverse HR-labeled data to train a cohort of resume-screening agents. By training 60 such agents with different models and data, we harness their crowd intelligence to optimize for three objectives: accuracy, fairness, and a balance of both. Furthermore, we develop a binary bias-detection model to visualize and analyze gender bias in both human and machine outputs. The results suggest that harnessing crowd intelligence using both accuracy and fairness objectives helps AI systems robustly output accurate and fair results. By contrast, a sole focus on accuracy may lead to severe fairness degradation, while, conversely, a sole focus on fairness leads to a relatively minor loss of accuracy. Our findings underscore the importance of balancing accuracy and fairness in AI-powered resume-screening systems to ensure equitable hiring outcomes and foster inclusive organizational development…(More)”
Problems of participatory processes in policymaking: a service design approach
Paper by Susana Díez-Calvo, Iván Lidón, Rubén Rebollar, Ignacio Gil-Pérez: “This study aims to identify and map the problems of participatory processes in policymaking through a Service Design approach….Fifteen problems of participatory processes in policymaking were identified, and some differences were observed in the perception of these problems between the stakeholders responsible for designing and implementing the participatory processes (backstage stakeholders) and those who are called upon to participate (frontstage stakeholders). The problems were found to occur at different stages of the service and to affect different stakeholders. A number of design actions were proposed to help mitigate these problems from a human-centred approach. These included process improvements, digital opportunities, new technologies and staff training, among others…(More)”.
The disparities and development trajectories of nations in achieving the sustainable development goals
Paper by Fengmei Ma, et al: “The Sustainable Development Goals (SDGs) provide a comprehensive framework for societal progress and planetary health. However, it remains unclear whether universal patterns exist in how nations pursue these goals and whether key development areas are being overlooked. Here, we apply the product space methodology, widely used in development economics, to construct an ‘SDG space of nations’. The SDG space models the relative performance and specialization patterns of 166 countries across 96 SDG indicators from 2000 to 2022. Our SDG space reveals a polarized global landscape, characterized by distinct groups of nations, each specializing in specific development indicators. Furthermore, we find that as countries improve their overall SDG scores, they tend to modify their sustainable development trajectories, pursuing different development objectives. Additionally, we identify orphaned SDG indicators — areas where certain country groups remain under-specialized. These patterns, and the SDG space more broadly, provide a high-resolution tool to understand and evaluate the progress and disparities of countries towards achieving the SDGs…(More)”
Big data for decision-making in public transport management: A comparison of different data sources
Paper by Valeria Maria Urbano, Marika Arena, and Giovanni Azzone: “The conventional data used to support public transport management have inherent constraints related to scalability, cost, and the potential to capture space and time variability. These limitations underscore the importance of exploring innovative data sources to complement more traditional ones.
For public transport operators, who are tasked with making pivotal decisions spanning planning, operation, and performance measurement, innovative data sources are a frontier that is still largely unexplored. To fill this gap, this study first establishes a framework for evaluating innovative data sources, highlighting the specific characteristics that data should have to support decision-making in the context of transportation management. Second, a comparative analysis is conducted, using empirical data collected from primary public transport operators in the Lombardy region, with the aim of understanding whether and to what extent different data sources meet the above requirements.
The findings of this study support transport operators in selecting data sources aligned with different decision-making domains, highlighting related benefits and challenges. This underscores the importance of integrating different data sources to exploit their complementarities…(More)”.
Overcoming challenges associated with broad sharing of human genomic data
Paper by Jonathan E. LoTempio Jr & Jonathan D. Moreno: “Since the Human Genome Project, the consensus position in genomics has been that data should be shared widely to achieve the greatest societal benefit. This position relies on imprecise definitions of the concept of ‘broad data sharing’. Accordingly, the implementation of data sharing varies among landmark genomic studies. In this Perspective, we identify definitions of broad that have been used interchangeably, despite their distinct implications. We further offer a framework with clarified concepts for genomic data sharing and probe six examples in genomics that produced public data. Finally, we articulate three challenges. First, we explore the need to reinterpret the limits of general research use data. Second, we consider the governance of public data deposition from extant samples. Third, we ask whether, in light of changing concepts of broad, participants should be encouraged to share their status as participants publicly or not. Each of these challenges is followed with recommendations…(More)”.
Smart cities: the data to decisions process
Paper by Eve Tsybina et al: “Smart cities improve citizen services by converting data into data-driven decisions. This conversion is not coincidental and depends on the underlying movement of information through four layers: devices, data communication and handling, operations, and planning and economics. Here we examine how this flow of information enables smartness in five major infrastructure sectors: transportation, energy, health, governance and municipal utilities. We show how success or failure within and between layers results in disparities in city smartness across different regions and sectors. Regions such as Europe and Asia exhibit higher levels of smartness compared to Africa and the USA. Furthermore, within one region, such as the USA or the Middle East, smarter cities manage the flow of information more efficiently. Sectors such as transportation and municipal utilities, characterized by extensive data, strong analytics and efficient information flow, tend to be smarter than healthcare and energy. The flow of information, however, generates risks associated with data collection and artificial intelligence deployment at each layer. We underscore the importance of seamless data transformation in achieving cost-effective and sustainable urban improvements and identify both supportive and impeding factors in the journey towards smarter cities…(More)”.
Towards Best Practices for Open Datasets for LLM Training
Paper by Stefan Baack et al: “Many AI companies are training their large language models (LLMs) on data without the permission of the copyright owners. The permissibility of doing so varies by jurisdiction: in countries like the EU and Japan, this is allowed under certain restrictions, while in the United States, the legal landscape is more ambiguous. Regardless of the legal status, concerns from creative producers have led to several high-profile copyright lawsuits, and the threat of litigation is commonly cited as a reason for the recent trend towards minimizing the information shared about training datasets by both corporate and public interest actors. This trend in limiting data information causes harm by hindering transparency, accountability, and innovation in the broader ecosystem by denying researchers, auditors, and impacted individuals access to the information needed to understand AI models.
While this could be mitigated by training language models on open access and public domain data, at the time of writing, there are no such models (trained at a meaningful scale) due to the substantial technical and sociological challenges in assembling the necessary corpus. These challenges include incomplete and unreliable metadata, the cost and complexity of digitizing physical records, and the diverse set of legal and technical skills required to ensure relevance and responsibility in a quickly changing landscape. Building towards a future where AI systems can be trained on openly licensed data that is responsibly curated and governed requires collaboration across legal, technical, and policy domains, along with investments in metadata standards, digitization, and fostering a culture of openness…(More)”.
Beware the Intention Economy: Collection and Commodification of Intent via Large Language Models
Article by Yaqub Chaudhary and Jonnie Penn: “The rapid proliferation of large language models (LLMs) invites the possibility of a new marketplace for behavioral and psychological data that signals intent. This brief article introduces some initial features of that emerging marketplace. We survey recent efforts by tech executives to position the capture, manipulation, and commodification of human intentionality as a lucrative parallel to—and viable extension of—the now-dominant attention economy, which has bent consumer, civic, and media norms around users’ finite attention spans since the 1990s. We call this follow-on the intention economy. We characterize it in two ways. First, as a competition, initially, between established tech players armed with the infrastructural and data capacities needed to vie for first-mover advantage on a new frontier of persuasive technologies. Second, as a commodification of hitherto unreachable levels of explicit and implicit data that signal intent, namely those signals borne of combining (a) hyper-personalized manipulation via LLM-based sycophancy, ingratiation, and emotional infiltration and (b) increasingly detailed categorization of online activity elicited through natural language.
This new dimension of automated persuasion draws on the unique capabilities of LLMs and generative AI more broadly, which intervene not only on what users want, but also, to cite Williams, “what they want to want” (Williams, 2018, p. 122). We demonstrate through a close reading of recent technical and critical literature (including unpublished papers from ArXiv) that such tools are already being explored to elicit, infer, collect, record, understand, forecast, and ultimately manipulate, modulate, and commodify human plans and purposes, both mundane (e.g., selecting a hotel) and profound (e.g., selecting a political candidate)…(More)”.
Governing artificial intelligence means governing data: (re)setting the agenda for data justice
Paper by Linnet Taylor, Siddharth Peter de Souza, Aaron Martin, and Joan López Solano: “The field of data justice has been evolving to take into account the role of data in powering the field of artificial intelligence (AI). In this paper we review the main conceptual bases for governing data and AI: the market-based approach, the personal–non-personal data distinction and strategic sovereignty. We then analyse how these are being operationalised into practical models for governance, including public data trusts, data cooperatives, personal data sovereignty, data collaboratives, data commons approaches and indigenous data sovereignty. We interrogate these models’ potential for just governance based on four benchmarks which we propose as a reformulation of the Data Justice governance agenda identified by Taylor in her 2017 framework. Re-situating data justice at the intersection of data and AI, these benchmarks focus on preserving and strengthening public infrastructures and public goods; inclusiveness; contestability and accountability; and global responsibility. We demonstrate how they can be used to test whether a governance approach will succeed in redistributing power, engaging with public concerns and creating a plural politics of AI…(More)”.
Boosting: Empowering Citizens with Behavioral Science
Paper by Stefan M. Herzog and Ralph Hertwig: “…Behavioral public policy came to the fore with the introduction of nudging, which aims to steer behavior while maintaining freedom of choice. Responding to critiques of nudging (e.g., that it does not promote agency and relies on benevolent choice architects), other behavioral policy approaches focus on empowering citizens. Here we review boosting, a behavioral policy approach that aims to foster people’s agency, self-control, and ability to make informed decisions. It is grounded in evidence from behavioral science showing that human decision making is not as notoriously flawed as the nudging approach assumes. We argue that addressing the challenges of our time—such as climate change, pandemics, and the threats to liberal democracies and human autonomy posed by digital technologies and choice architectures—calls for fostering capable and engaged citizens as a first line of response to complement slower, systemic approaches…(More)”.