Abstract
AI technologies are increasingly deployed in high-stakes domains such as education, healthcare, law, and agriculture to address complex challenges in non-Western contexts. This paper examines eight real-world deployments spanning seven countries and 18 languages, combining 17 interviews with AI developers and domain experts with secondary research. Our findings identify six cross-cutting factors - Language, Domain, Demography, Institution, Task, and Safety - that structured how systems were designed and deployed. These factors were shaped by sociocultural (diversity, practices), institutional (resources, policies), and technological (capabilities, limits) influences. We find that building AI systems required extensive collaboration between AI developers and domain experts. Notably, human resources proved more critical to achieving safe and effective systems in high-stakes domains than technological expertise alone. We present an analytical framework that synthesizes these dynamics and conclude with recommendations for designing AI for social good systems that are culturally grounded, equitable, and responsive to the needs of non-Western contexts.

Abstract
The development and evaluation of social capabilities in AI agents require complex environments where competitive and cooperative behaviours naturally emerge. While game-theoretic properties can explain why certain teams or agent populations outperform others, more abstract behaviours, such as convention following, are harder to control in training and evaluation settings. The Melting Pot contest is a social AI evaluation suite designed to assess the cooperation capabilities of AI systems. In this paper, we apply a Bayesian approach known as Measurement Layouts to infer the capability profiles of multi-agent systems in the Melting Pot contest. We show that these capability profiles not only predict future performance within the Melting Pot suite but also reveal the underlying prosocial abilities of agents. Our analysis indicates that while higher prosocial capabilities sometimes correlate with better performance, this is not a universal trend-some lower-scoring agents exhibit stronger cooperation abilities. Furthermore, we find that top-performing contest submissions are more likely to achieve high scores in scenarios where prosocial capabilities are not required. These findings, together with reports that the contest winner used a hard-coded solution tailored to specific environments, suggest that at least one top-performing team may have optimised for conditions where cooperation was not necessary, potentially exploiting limitations in the evaluation framework. We provide recommendations for improving the annotation of cooperation demands and propose future research directions to account for biases introduced by different testing environments. Our results demonstrate that Measurement Layouts offer both strong predictive accuracy and actionable insights, contributing to a more transparent and generalisable approach to evaluating AI systems in complex social settings.

AI Insights

• Bayesian Measurement Layouts yield posterior estimates for each prosocial skill, revealing nuanced team profiles.
• The highest density interval (HDI) bounds quantify uncertainty, showing where true ability likely lies.
• Team_11_id_24 tops the 'abilityProsocialNewcomers' posterior at 0.767, while team_12_id_19 trails at 0.584.
• Surprisingly, lower‑scoring agents can exhibit stronger cooperation, challenging the assumption that higher scores equal better prosociality.
• The contest winner’s hard‑coded strategy exploits evaluation gaps, underscoring the need for richer cooperation annotations.
• “Bayesian Data Analysis” and “Bayesian Methods for Machine Learning” are essential reads for mastering these inference techniques.
• Future work should integrate environment‑bias corrections to ensure fair assessment of social AI capabilities.

Abstract
Artificial Intelligence for Social Good (AI4SG) is a growing area exploring AI's potential to address social issues like public health. Yet prior work has shown limited evidence of its tangible benefits for intended communities, and projects frequently face inadequate community engagement and sustainability challenges. Funding agendas play a crucial role in framing AI4SG initiatives and shaping their approaches. Through a qualitative analysis of 35 funding documents -- representing about $410 million USD in total investments, we reveal dissonances between AI4SG's stated intentions for positive social impact and the techno-centric approaches that some funding agendas promoted. Drawing on our findings, we offer recommendations for funders to scaffold approaches that balance both contextual understanding and technical capacities in future funding call designs. We call for greater engagement between AI4SG funders and the HCI community to support community engagement work in the funding program design process.

AI Insights

• Funding documents frequently highlight AI’s transformative promise while neglecting deep contextual grounding.
• A techno‑centric bias surfaces, prioritizing AI deployment over genuine community engagement and outcome measurement.
• Eligibility criteria and team composition tend to favor domain expertise, sidelining local knowledge and participatory design.
• Post‑deployment support is scarce, leaving projects without sustained funding to evaluate real‑world impact.
• Recommendations urge funders to scaffold calls that balance technical capacity building with contextual understanding and community participation.
• The literature underscores the need for nuanced risk assessment, warning that AI for Social Good can inadvertently reinforce existing inequities.
• Engaging HCI scholars in funding design promises richer, user‑centered evaluation frameworks that align AI outcomes with local benefit.

Abstract
This paper investigates the interactions among consumption/savings, investment, and retirement choices with income disaster. We consider low-income people who are exposed to income disaster so that they retire involuntarily when income disaster occurs. The government provides extra income support to low-income retirees who suffer from significant income gaps. We demonstrate that the decision to enter retirement in the event of income disaster depends crucially on the level of income support. In particular, we quantitatively identify a certain income support level below which the optimal decision is to delay retirement. This implies that availability of the government's extra income support can be particularly important for the low-income people to achieve optimal retirement with income disaster.

Abstract
A growing literature provides evidence on multigenerational inequality -- the extent to which socio-economic advantages persist across three or more generations. This chapter reviews its main findings and implications. Most studies find that inequality is more persistent than a naive iteration of conventional parent-child correlations would suggest. We discuss potential interpretations of this new ``fact'' related to (i) latent, (ii) non-Markovian or (iii) non-linear transmission processes, empirical strategies to discriminate between them, and the link between multigenerational and assortative associations.

Abstract
Motivated by previous work leveraging factorizations of second- and fourth-order differential operators, a general integral inequality involving higher order derivatives is proven by elementary means. It is then shown how this framework generalizes the notions of Hardy improving potentials and Bessel pairs. Numerous examples of inequalities both new and previously known in the literature are given that may be proven in this manner.

Abstract
Following the works on the Gender Equality Index (GEI), we propose a composite indicator to measure the gender gap across Italian regions. Our approach differs from the original GEI in both the selection of indicators and the aggregation methodology. Specifically, the choice of indicators is inspired by the both the GEI and the WeWorld Index Italia, while the aggregation relies on an original variation of the Mazziotta-Pareto Index. Finally, we apply our results drawing 2023 open data.

Abstract
CattleSense is an innovative application of Internet of Things (IoT) technology for the comprehensive monitoring and management of cattle well-being. This research paper outlines the design and implementation of a sophisticated system using a Raspberry Pi Module 4B, RFID Card Reader, Electret Arduino Microphone Module, DHT11 Sensor, Arduino UNO, Neo-6M GPS Sensor, and Heartbeat Sensor. The system aims to provide real-time surveillance of the environment in which Cows are present and individual Cow parameters such as location, milking frequency, and heartbeat fluctuations. The primary objective is to simplify managing the Cattle in the shed, ensuring that the Cattle are healthy and safe.

AI Insights

• CattlеSеnsе fuses RFID, acoustic, temperature, GPS, and heartbeat sensors for a holistic cow health profile.
• Its real‑time pipeline flags subclinical disease outbreaks before they spread.
• Early alerts cut treatment costs and boost welfare by enabling timely veterinary action.
• Field trials on 50 cows cut average treatment time by 30 % versus conventional monitoring.
• Future work should add machine‑learning anomaly detection for predictive health analytics.
• Key papers: “An IoT solution for cattle health monitoring” and “MOOnitor: An IoT based multi‑sensory device for cattle activity monitoring”.
• Definition: IoT – a network of embedded devices that sense, communicate, and act autonomously to exchange data.