JOURNAL OF SOCIAL SCIENCES DEVELOPMENT

THE HUMAN SIDE OF AI: HOW ETHICAL LEADERSHIP SHAPES EMPLOYEE OUTCOMES IN HIGHER EDUCATION

Dr. Sonia Sethi, Dr. Hina Saleem, Dr. Syed Noman Mustafa — Wed, 02 Sep 2026 00:00:00 +0000

Artificial intelligence (AI) is increasingly transforming higher education through applications in teaching, research, assessment, administration, student support, and institutional decision-making. Although AI offers substantial opportunities for efficiency and innovation, its implementation also creates organisational and human challenges related to transparency, fairness, accountability, professional autonomy, job uncertainty, and employee well-being. This study examines the role of ethical leadership in shaping employee outcomes within the evolving context of AI-enabled higher education. A quantitative, cross-sectional research design was adopted, and data were collected from academic and administrative employees in higher education institutions using a structured questionnaire based on a five-point Likert scale. The study examined ethical leadership in relation to four employee outcomes: job satisfaction, organisational commitment, employee engagement, and employee performance. Partial Least Squares Structural Equation Modelling (PLS-SEM) using SmartPLS was employed to assess the measurement and structural models and test the proposed relationships. The findings indicate that ethical leadership has positive relationships with job satisfaction, organisational commitment, employee engagement, and employee performance. The results suggest that fairness, integrity, transparency, accountability, and employee-oriented leadership can strengthen positive employee responses during periods of technological and organisational change. The study further highlights that successful AI integration in higher education requires more than technological infrastructure and digital competencies. Human-centred leadership, employee participation, transparent communication, appropriate training, ethical governance, and organisational support are essential for responsible and sustainable AI transformation. The study contributes to the emerging literature by connecting ethical leadership and employee outcomes with the human dimensions of AI-enabled transformation in higher education, particularly within the developing higher education context.

BEYOND THE BLUNT VERDICT: A FOUR-DIMENSION DIAGNOSTIC FRAMEWORK FOR INTELLIGENCE WARNING FAILURE IN BALOCHISTAN

Muhammad Faizan Asghar — Thu, 03 Sep 2026 00:00:00 +0000

The concept of intelligence failure is a constant refrain in Pakistan's security establishment when it comes to the various rounds of shooting sprees that have been going on in Balochistan. This is a way of hiding more than showing. It cannot tell if someone interpreted a warning correctly but it did not pass through to a decision-maker, or it was meant to be handed out to a decision-maker but passed no one up the chain of command, or if it was meant to be a warning but wasn't recognized as such in the first place. The paper draws on the recent literature on intelligence studies for developing a four-dimensional framework for the analysis of warning, comprising analysis, process, politics, and tribal integration, and extends the framework by adding a tribal and administrative geography dimension for Balochistan. This framework is used for the ongoing insurgent violence by Balochs that has brought terror to the multi-district attack in January–February 2026; the hijacking of Jaffar Express on March 2025 and the live tempo of attacks through mid-2026. The analysis reveals a less effective warning chain – tactical, short horizon analysis is often sufficient, as the process is fragmented within a crowded intelligence bureaucracy, political unwillingness to rethink resource allocation in favor of the central warning chain, and an unreliable tribal information network results in dampening of warning value if there is any. Finally, the paper offers a suite of institutional (as opposed to operational) recommendations to help reduce the disconnect between what the state knows and what the state can do.

WEATHER FORECASTS AS A CREDENCE GOOD: VERIFIED ACCURACY, FARMER BELIEF AND THE DEMAND FOR AGROMETEOROLOGICAL INFORMATION IN PAKISTAN

Abdul Basit, Nasir Iqbal — Thu, 03 Sep 2026 00:00:00 +0000

A farmer who acts on a weather advisory and still harvests a poor crop cannot tell a bad forecast from bad luck. Forecast quality is consumed jointly with soil, seed, pests and the weather draw, so consumption does not reveal it. Almost every study of weather-information uptake observes what farmers believe about forecast quality but not what that quality is. This paper measures both sides at the same place and the same time. Verified day-ahead accuracy is constructed for 89 Pakistan Meteorological Department stations from 34,610 scored station-days and matched to the perceived accuracy reported by 689 farmers. The level comparison depends on which verification score is used. Against percentage correct, mean belief of 0.692 sits 0.19 below a verified 0.883, but percentage correct is beaten on this panel by a no-skill climatological rule, and the sign of the gap reverses against the hit rate on rain days. What does not depend on the score is the decoupling. Across seven verification metrics the correlation between belief and verified quality never exceeds 0.112 in magnitude at station level or 0.056 at farmer level. Demand follows belief alone. Conditional on the perception, verified accuracy has no detectable effect on willingness to pay or reported outcomes. A learning model with misattribution, at calibrated coding thresholds, reproduces about two thirds of the level gap measured by percentage correct. Its policy ranking places de-biasing extension and localised advisories far above subsidised access.

MIGRATIONS AND ECONOMY: UNDERSTANDING DISPLACEMENT DRIVEN ECONOMIES

Faraz Ali, Amir Raza — Fri, 04 Sep 2026 00:00:00 +0000

Displacement creates complex social and economic challenges between displaced and host communities but also new opportunities for integration. This study examines the economic participation of displaced populations in Tehsil Razar, District Swabi, Khyber Pakhtunkhwa, Pakistan, arguing that it has evolved into a displacement-driven economy. Based on 30 semi-structured interviews with traders, farmers, landowners, elders, and officials, findings show Afghan refugees dominate trade while internally displaced persons sustain agricultural labor. Despite legal and social barriers, displaced groups are vital to the local economy. The paper introduces the concept of a “displacement-led economy” with relevance to Uganda, Lebanon, Jordan, and Europe.