The AI Shock in Higher Education: Exploring Digital Precarity, Pedagogical Authority, and Social Reproduction in District Hyderabad Colleges through Mixed-Methods and SEM Analysis

Copyright © ISRG Publishers. All rights Reserved. DOI: 10.5281/zenodo.17596577 165 ISRG PUBLISHERS Abbreviated Key Title: ISRG J Arts Humanit Soc Sci ISSN: 2583-7672 (Online) Journal homepage: https://isrgpublishers.com/isrgjahss Volume– III Issue -VI (November - December) 2025 Frequency: Bimonthly The AI Shock in Higher Education: Exploring Digital Precarity, Pedagogical Authority, and Social Reproduction in District Hyderabad Colleges through MixedMethods and SEM Analysis Muhmmad Asif1, Huma Shareef2*, Muhammad Javed Sheikh3, Faisal Karim Memon4, Farhan Akhtar5, Abdul Rasool Khoso6 1, 5, 6 Department of Sociology, School of Public Administration, Hohai University, Nanjing, China. 2, 3, 4 Department of Rural Sociology, FASS, Sindh Agriculture University, Tandojam, Pakistan. | Received: 07.11.2025 | Accepted: 11.11.2025 | Published: 13.11.2025 *Corresponding author: Huma Shareef Department of Sociology, School of Public Administration, Hohai University, Nanjing, China. [email protected] Abstract The present study explores the sociological consequences of the addition of artificial intelligence in Pakistan's higher education, theorizing the phenomenon as an AI shock that researches academic labor, teaching authority, and inequality within the global south. Using Bourdieu's capital theory, Kalleberg's precarity framework, and Giddens' structuration theory, this research assumes a progressive explanatory mixed method design combining quantitative descriptive and regression analysis by using SPSS, further inferential methods like correlation, heat map, regression coefficient, and SEM models analyses by using Python, as well as the qualitative thematic analysis of teacher interviews. Data were collected from 300 educators at public and private colleges in Tandojam. This indicates that while AI integration moderately enhances digital capital and pedagogical authority, it also increases job instability and institutional dependency. Regression and SEM results show that pedagogical authority (beta = 0.37) and digital capital (beta = 0.38) have substantial adverse effects. Qualitative models confirm that teachers experience AI as both an empowering tool and a source of fear, shaped by institutional pressure, gendered workload, and irregular access to training. These findings suggest that AI-driven reforms further reproduce existing orders under conditions of policy-driven digitalization, a process that resonates with Bourdieu and Passeron's 1990 social reproduction theory. The study examines these dynamics through the prism of the Sustainable Development Goals (SDGs 4, 8, and 10), highlighting that quality education, decent work, and reduced inequalities will require reorienting AI adoption toward equitable labor practices and ethical governance. The present study concludes by conceiving digital precarity as a defining feature within the outlines of the 21st-century academic field and calls for a human-centered digital transition that foregrounds teacher agency, institutional justice, and social sustainability. Keywords: Artificial Intelligence, Digital Precarity and Capital, Pedagogical Authority, Sociology of Education, Social Reproduction Theory and SDGs, Higher Education, Pakistan. Copyright © ISRG Publishers. All rights Reserved. DOI: 10.5281/zenodo.17596577 166 Introduction The global education landscape is undergoing a rapid transformation, driven by advancesn artificial intelligence (AI) (Roy & Paul, 2023). Advances in education technology, such as large language models like ChatGPT and Gemini, and AI that grades, have evolved from theory to actual teaching practice and are now part of educational policy. (Rane, 2024). This development, known as the AI shock, is a disruptive force that extends beyond educational efficiency and academic integrity. The sociological implications of AI Shock represent a substantial reordering of the education field, with the potential to disentrance power relations, professional identities, and social reproduction. (Natsiopoulou, 2025). While previous articles focused on student experiences, ethics, and future assessment strategies, another important aspect of this technological change that deserves attention is its impact on teachers, especially within the oftencomplex, hierarchical context of the Global South. In Pakistan, however, this mechanization is not a naturally occurring adoption from the bottom up but increasingly a top-down policy construct. The government's "Digital Pakistan" vision and the Higher Education Commission's directives are both proactive efforts to promote technology integration in higher education, presenting institutions with a possible ecosystem in which legitimacy stems from policy compliance. (Ahmad, et al., 2025; Ying, et al., 2024). This swift integration, which is driven by policy, presents an issue. It is argued that it poses a risk of exacerbating teachers' precarious conditions and systematically altering traditional forms of pedagogical authority. Precarity, with its state of existential and economic insecurity and lack of control over one's work (Castro, 2022). It is an established characteristic of the educational labor markets in the global south. However, the AI shock introduces a new digital layer to its precarity. Teachers, particularly at resource-constrained or non-elite institutions, face dual pressures: to master not only opaque, rapidly evolving AI tools but also to maintain their professional legitimacy in the classroom. This shifting dynamic creates a new axis of inequality among the professoriate, in which a teacher's worth is tied to their digital capital. (Pitzalis & Porcu, 2024). Potentially devaluing deep content knowledge and experience-based wisdom as a component of pedagogy. Research Gap This brings out a significant gap in the existing research. Until now, the majority of the literature on AI in education has been found in the Western, technologically dense ecosystem and tends to be student-centered in its use of learning outcomes. The lack of sociological inquiry into teachers' labor is stark, not to mention areas like South Asia, where educational systems are characterised by strong hierarchies, bureaucracies, and social inequalities (Zulfiqar & Prasad, 2021). Indeed, little is known about how this change is perceived by college teachers in Pakistan, how it influences their perceptions of professional autonomy and selfefficacy, and the strategies they use to manage the new demands imposed on them. Such an absence in the literature is critical, as it fails to acknowledge how global technological forces manifest through local institutionalized structures and teacher agency, thereby perhaps exacerbating existing social inequities. Theories Applied in The Study: The presented work is based on Bourdieu's theories of the field of capital and habitus, in which the college setting is viewed as a dynamic field shaped by me (Schirone, 2023). The change can focus on emerging forms of digital capital and lead to a crisis of habitus among teachers whose careers were formed before the AI era. Further, the social reproduction theory examines how the unequal assimilation of AI tools among teachers across different institutional contexts perpetuates current inequalities in education (Ikpuri, 2023). The study is relevant to the sociology of education because it combines technological transformation with traditional theories of labor and capital to develop the concept of digital precarity to comprehend modern academic work better. It also provides valuable qualitative and quantitative data in the underresearched setting of Pakistani higher education, illuminating the challenges of agency and the constraints on teachers. On the political side, the research will seek to offer evidence-based recommendations to policymakers and institutional leaders in Pakistan and other such settings, to promote fair AI integration policies that will resolve, rather than exacerbate, problems of precarity and inequality. By framing the AI shock as a significant sociological event rather than a technological advancement, this study aims to highlight the crucial role of human labor in transforming education. Literature review The adoption of AI in education should not be interpreted as a nonpartisan, non-technical, or technical advancement, but as a sociotechnical process rooted in current power systems, labor relations, and institutional hierarchies. It is based on this that this review synthesizes literature on three critical areas to situate the current study: that firstly, the developing body of scholarship on technology and teacher labor identifies a gap in comprehending the role of AI in promoting precarity, secondly that theories of social reproduction are explored and applied to the digital realm, and finally, the body of work is contextualized to Pakistan higher education, which is a unique and under-studied field. Technology and Teacher Labor: Tool to Threat. This has been a long scholarly issue on the connection between educational technology and teacher labor. Contrary to the technology-optimistic discourses that saw technology as a tool for teacher empowerment, promising to lighten the administrative load and enable teachers to instruct more personally. (Birbirso, 2013). The possibility of technology de-skilling, intensifying, and controlling teaching work has remained in the limelight through a more critical sociological tradition. (Beech, 2021). The present AI revolution puts this strain into sharp focus. One of the studies suggests that AI can increase the capacity of teachers, and the results of the research in the West show the possibility of automated grading and learning analytics to enable teachers to concentrate on more complicated pedagogical relation (Salas-Pilco, et al., 2022). Still, this assumption overlooks the amount of work that teachers have to invest in learning and using the new technologies. However, on the contrary, a lot of the literature is pointing at technology-enabled intensification, i.e., implementing new tools that do not help to reduce the already existing workloads of teachers (Pappa, et al., 2024) The most important thing that is lacking in this discourse is the peculiarity of AI. In contrast to previous educational technologies, AI systems often become black boxes, making independent decisions that were previously assigned to teachers. This change has a direct impact on the pedagogical authority, which is described as the authoritative right to make decisions about the curriculum and the instruction (Gerrard & Farrell, 2014). Copyright © ISRG Publishers. All rights Reserved. DOI: 10.5281/zenodo.17596577 167 This loss of power is connected with the notion of precarity. Such a state of persistent insecurity and lack of control, as (Kalleberg & Vallas, 2017) Defines precarity as a descriptive term used in reference to the academic profession. Digital output-based performance metrics increase faculty insecurity, according to research conducted in Singaporean universities. (Baildon, 2025). Nonetheless, there is not much research that specifically examines the connection between the integration of AI and the precarity of teachers in a situation where job security may already be precarious. Aspects of attitudes and adoption, including (Mancaniello & Lavanga, 2024), have received the most attention in the international literature, without regard for how the pressure to use AI is transforming the nature of teaching as a profession. This paper directly examines whether AI can be regarded as an augmenting tool or as a threat, thereby making professions less secure and reducing their autonomy in the Pakistani college setting. Digital Age Social Reproduction: Redescribing Inequality. Based on the arguments of (Bourdieu, 1990), social reproduction implies that the educational system reinforces and legitimize social inequalities between and among generations due to the unequal allocation of cultural, social, and economic capital. In the era of digital technology, this framework needs a crucial update, which puts the concept of digital capital into the picture as a critical determinant of the inequality picture. Digital capital is the resources acquired through exposure to technology, the skills of its use, and the social networks that enable such interactions, and is becoming critical to meaningful involvement in social and economic life. (Merisalo & Makkonen, 2022). As the first-level digital divide, which focuses on physical access to technology, this has been extensively analyzed. (Van Deursen & Van Dijk, 2019), the second-level digital divide which deals with skills inequality and usage is now coming into the spotlight. Moreover, in this analysis, it is assumed that there is a third-level AI divide, suggesting the possibility of using high-level AI tools to achieve academic and professional success. The studies show that students with higher incomes are more proficient at using AI to solve complex problems, such as coding and data analysis. Conversely, underprivileged individuals in society are likely to use it to perform simpler tasks, which only widens the achievement gap. (Rothstein, 2013). However, this reproduction logic applies to teachers, as well as their institutions. Elite colleges with sufficient resources are able to invest in AI infrastructure, training, and provision to endow their faculty with high digital capital. (Chakravarty, 2022). Such teachers would be better positioned to ensure that AI is seamlessly integrated into their teaching practice, thereby legitimizing their institution as elite and their own professional authority as teachers. On the other hand, teachers in under-resourced colleges are not provided with such resources and support and may lag behind, thus creating a further hierarchy within the system based on the technological capabilities of the college. (Akyeampong, 2022). Where there is a bit of scholarship, as in Entrich (2020), that demonstrates how the shadow education system recreates inequality, there is nevertheless a fundamental gap in how internal digital stratification of the formal education system can play into such processes. This paper directly examines how the unequal allocation of digital capital among educators across the various institutional forms in Pakistan constitutes a new form of social reproduction, and how these implications may have far-reaching consequences for the establishment of a two-level exemplar of pedagogical practice. The Pakistani Higher Education Context: A Perfect Storm Inherent to the Pakistani higher education system, the interactions among technology, labor, and inequality take a particular form that makes the place one of the most interesting for sociological analysis. A clear-cut dualism defines the system: a tiny segment of elite, well-funded private and public universities partners with a huge public and low-tier private college sector that is often suffering from a lack of resources, slow-moving bureaucracy, and political interference. (Ajibade & Ibietan, 2016; Chen & Khoso, 2025). There is certainly a lot of coercive pressure in the policy environment. The HEC's push for digitalization is well-intentioned but often presents itself as a top-down command without the corresponding investment in infrastructure or professional development (Nawaz, 2012). This thus leads to a classic situation of "coercive isomorphism" from new institutional theory, where organizations adopt policies merely to gain legitimacy without substantial implementation. The situation for college teachers is a profound double bind: they are required to use technology, but the resources they are given are so inadequate that they are likely to fall victim to the digital precarity that this research aims to reveal (Stewart-Robertson, 2022). In the past, educational research in Pakistan has been focused almost exclusively on issues such as access to education, gender equality, and curriculum reform. (Sain, 2023; Durrani & Halai, 2018). Nevertheless, the topic of how the digital policy affects the academic labor force sociologically is almost entirely missed in the literature. The scant studies that touch on the topic of technology, such as the one by Khalil (2021) on e-learning readiness, regard the issue from a technocratic, capacity-building standpoint and hence avoid the central questions of power, labor, and inequality. They do not inquire about the type of interactions taking place between global technological forces like AI and Paki Methodology Research Design This study employed a mixed-methods research design, combining quantitative and qualitative approaches to explore how artificial Intelligence (AI) influences teachers’ work, authority, and job security in higher education. The quantitative phase focused on measuring relationships among key constructs using statistical models. In contrast, the qualitative phase provides deeper insights into teachers’ experiences and interpretations of AI in their professional contexts. Study Area Figure 1: Study Area Map This research was conducted in District Hyderabad, Sindh Province, Pakistan, focusing specifically on Tandojam, a suburban Copyright © ISRG Publishers. All rights Reserved. DOI: 10.5281/zenodo.17596577 168 educational hub known for its mix of public and private colleges. The area represents a balanced blend of traditional and emerging digital pedagogies, making it ideal for studying how artificial intelligence reshapes teaching practices. Hyderabad’s colleges serve diverse socioeconomic groups, allowing the study to capture variations in institutional resources, teacher experiences, and exposure to digital technologies. Sample size and sampling This research contacted both public and private teachers in Schools and colleges in Tandojam. Approximately 300 teachers were asked to complete the questionnaire, and 20 were selected for interviews from different schools and with diverse experiences. Quantitative data were collected through stratified random sampling, while purposive sampling was used for in-depth interviews. Data Collection Quantitative data were collected through a structured questionnaire measuring AI integration, digital capital, pedagogical authority, job precarity, and performance. Some open-ended questions were asked of teachers about their perceptions of AI adoption, institutional challenges, how it is changing their teaching and workloads, and how they are coping at school. Data Analysis SPSS and Python (v3.11) software were used for quantitative and qualitative analysis, applying descriptive statistics and regression in SPSS 25, and correlation, regression Coefficient Graph, and Structural Equation Modeling (SEM) to test hypothesized relationships among variables were performed using SciPy, Pingouin Stats, and Semopy in Python. Matplotlib and Seaborn were used to create easy-to-read charts and graphs. Nevertheless, Qualitative data were analyzed using Python, NLTK, and spaCy to identify common themes, including fear of technology, inequalities in access to technology, school pressure, and changes in thinking about technology. This gave a story to accompany the figures about how teachers are adjusting to AI changes. How It Fits with SDGs Goals The provided research can be connected with the United Nations Sustainable Development Goals (SDGs):  SDG 4 (Good Education): Promoting technology literacy.  SDG 8 (Good Jobs): Peeking at occupational safety.  SDG 10 (Fairness): Showing who are tech-haves and who are not.  SDG 17 (Working Together): The request to other people in other fields to cooperate in advancing education. With these objectives in mind, this study strives to help normalize the situation in teaching and ensure that technology is used effectively in education. Results Objective: 1 Examine how AI integration influences job precarity, professional autonomy, and pedagogical authority Table 1: Descriptive statistics of the variables Items Mean SD Min Max AI Integration 3.11 1.114195 1 5 Digital Capital 3.2 1.092439 1 5 Job Precarity 3.33 1.14857 1 5 Pedagogical Authority 2.87 1.063394 1 5 Performance 2.80 1.04664 1 5 Table 1: The descriptive result reveals moderate mean values for AI Integration (3.11), Digital Capital (3.24), and Pedagogical Authority (3.33). Nonetheless, Job Precarity (2.87) is slightly lower and indicates an average use of digital tools, yet with stress regarding professional well-being. Theory link: A. L. Kalleberg's precarity theory (2018) explains that current digital labor creates new uncertainties due to the pressure of managers or policymakers. In this case, AI establishes two situations: institutional control and empowerment by autonomy. Figure 2: Relationship between the Variables The heatmap shows strong positive correlations between AI Integration and Digital Capital (r = 0.59) and moderate positive correlations between AI Integration and Pedagogical Authority (r = 0.34). Altogether, the educators who use AI tools will gain more digital competence and self-efficacy in their classroom behavior. Nevertheless, the results reveal low or negative correlations between Job Precarity and the constructs (r = -0.25 to -0.16), or, more straightforwardly, that fear of losing the job reduces the gains of technological involvement. Comprehensively, this validates A. L. Kalleberg's precarity theory. (2018) that shows how institutional instability is an intermediate that mediates the positive opportunities provided by AI. Objective 2: Investigate the connection between AI adoption, digital capital, and teacher performance. Table 2: Multiple Regression Influence on Teacher Performance Predictor Variable β (Standardized) t-value Sig. (p) AI Integration 0.11 1.94 0.054 Digital Capital 0.20 3.02 0.003 Pedagogical Authority 0.37 5.66 0.000 Job Precarity -0.38 -6.24 0.000 Model Summary: R² = 0.61, Adjusted R² = 0.59, F(4, 216) = 35.7, p < 0.001 In Table 2, the regression analysis findings reveal that Pedagogical Authority (b = 0.37, p < .001) and Digital Capital (b = 0.20, p = Copyright © ISRG Publishers. All rights Reserved. DOI: 10.5281/zenodo.17596577 169 .003) have positive effects on teacher performance. In contrast, Job Precarity (b = -0.38, p < .001) harms teacher performance. The incremental effect of AI Integration (b = 0.11, p = .05) is present. The model predicted performance at very high rates, since it explained 61 percent of the variance. This proves that authority and digital capital are two significant forms of professional capital that modern-day teaching requires an answer to Bourdieu (1990) theory of capital conversion. Precarity opposes the stability and motivation of institutions. Figure 3: Predicting Regression Coefficient Betas According to the regression coefficients, Pedagogical Authority (b = +0.37) and Digital Capital (b = +0.19) are the two best positive predictors of teacher performance. AI Integration is positively related to performance (b = +0.11), and Job Precarity negatively influences job performance (b = -0.38). This also supports Bourdieu's (1990) ideas on capital conversion, in that teachers can convert digital capital competencies into symbolic authority, which together represent higher performance in their professional duties. On the other hand, at a micro level, if the teacher is working within structural insecurity due to precarity, that process may be hampered, and the overall institutional aspects of the digitalization experience may decrease for schools. Figure 4: The SEM model illustrates a strong mediated relationship SEM Path Model • AI Integration → Digital Capital (β = 0.59) • Digital Capital → Pedagogical Authority (β = 0.49) • Pedagogical Authority → Performance (β = 0.37) • Job Precarity → Performance (β = –0.38) This suggests that AI's positive effects on performance work indirectly through the use of skills and greater authority, while sources of insecurity continue to undermine them. The model puts Bourdieu’s (1990) field-habitus theory into practice, with teachers reorienting themselves in a changing digital field and, depending on the context of institutions, changing AI from a threat to a resource. Figure 5: Faculty Effectiveness Performance SEM Model The Structural Equation Model (SEM) posits that faculty effectiveness enhances student performance directly but also indirectly through the two mediating constructs, communication Copyright © ISRG Publishers. All rights Reserved. DOI: 10.5281/zenodo.17596577 170 and social behavior. Specifically, mentorship (β = 0.68) and time management (β = 0.42) are strong predictors of effective communication with students. Meanwhile, punctuality (β = 0.56) and character (β = 0.61) are strong predictors of a faculty member's social behavior with students. Communication (β = 0.73) and social behavior (β = 0.58) are both strong, positive predictors of performance, suggesting that students' success is contingent on the cognitive, communicative, and moral–behavioral dimensions of faculty engagement. These findings align with Bandura's Social Learning Theory, which suggests that students model teachers' communicative and ethical behavior, and with Bourdieu’s Capital Theory (1986), Cultural Capital Theory, which posits that faculty embody values and behaviors that are transferred to learners as cultural context. Hence, the SEM model shows that mentorship, discipline, and moral character promote communication and social behavior in an academic context, which, in turn, increase performance levels, a modeling occurrence that demonstrates how professional competence and moral capital coalesce to foster educational quality as understood within the sociology of education frame. Objective 3: Qualitative Investigate how educators deal with, oppose, or adjust to AI technologies. Figure 6: Conceptual Map of SDG-Related Theme This map depicts two forces at work at the same time, empowering Mechanisms (Green and Bluepaths): AI → Digital Capital → Authority → Performance (supporting SDG 4 & 8). Red pathways (restrictive mechanisms): precarious employment, institutional coercion, and technological anxiety (threat to SDG 10). When educators referred to AI as a tool of hope and fear, they indicated unequal access and top-down pressure, which is reflective of the social reproduction theory of Bourdieu and Passeron's (1990), which indicates that without institutional equity to support digital transformation, we risk reproducing systemic inequities. Figure 7: Qualitative Themes in a Hierarchical Tree Model The hierarchical model illustrates the complexity of how the AI Shock in Education is instrumentalized, enacting multifaceted change across the pedagogical, institutional, and emotional aspects of the teacher experience. The tensions educators experience due to the adaptation of reforms to classroom teaching practice, as documented in the technological Anxiety and Digital Precarity and Institutional Coercion and Inequality of Access, are a consequence of Kalleberg's (2018). Precarity Theory, which holds that technological change engenders new insecurities within professional labour. Emotional labour, as evidenced by exhaustion, fear of displacement, and unequal professional development, helps us understand how teachers reconcile uncertainty and the mechanisms that create additional structural pressures. Moving downward, Digital Capital as Empowerment reflects Bourdieu’s Capital Theory (1986). Teachers who develop AI competence accumulate digital capital, gaining symbolic and professional authority. Re-Negotiating Pedagogy and Authority, on the other hand, echoes Giddens' Structuration Theory (1984), which holds that institutional structures and human agency constantly coconstruct social reality by showing how educators reshape their roles and instructional identities. In line with theory of power/knowledge Gordon (1980), the middle tier, Labour, Power & Sustainability Nexus, integrates gender inequality, policy pressure, and ethical guidance. It focuses on how structural power and policy environments mediate teacher agency. The digital order has decentralized and contested authority in education. At the bottom, linking to SDG 4 (Quality Education), SDG 8 (Decent Copyright © ISRG Publishers. All rights Reserved. DOI: 10.5281/zenodo.17596577 171 Work), and SDG 10 (Reduced Inequalities) proves that the sustainable digital transformation will have to reconcile innovation and fairness. Skill development of teachers and their new roles (SDG 4) will be conditioned by decent work environments (SDG 8), while reducing inequities and gender discrimination will guarantee that inequalities are reduced (SDG 10). Overall, the model suggests that the transformation of the field of education with AI is not just a technological challenge but also a big sociological one - this is a wholesome reset of habitus and capital in the era of uncertainty, where the triad of empowerment, equity, and ethical labor is the pillar of sustainable digital education. Figure 8: Themes World cloud Figure 8 presents a cluster map that unites all the major concepts and places AI Shock at the center of a network connecting empowerment (including digital capital and authority) and constraints (including anxiety and coercion). The help sociological model of digital precarity and professional adaptation to reality links hard evidence with theory. Theoretical integration in interpreting this question has the cluster map illuminating the sociological relationships between teachers and students, changing the AI shock in education. It correlates five qualitative themes with Sustainable Development Goals (SDGs 4, 8, and 10). The Technological Anxiety Digital Precarity is a theme in which the author exposes teachers' emotional and professional vulnerabilities in response to the forces of automation. It aligns with Precarity Theory (2018) by Kalleberg, who writes about the increasing insecurity in professional labor markets due to the rise of new technologies, which poses a challenge to SDG 8 (Decent Work). On the other hand, Digital Capital as Empowerment is a repetition of Bourdieu’s (1986) Capital Theory, focusing on the beneficial side of digital resources in increasing teachers' power and abilities. Tech skills are also a new form of digital asset that enhances educators' credibility and teaching skills, which supports the Quality Education goal. The Institutional Coercion and Inequality theme, which is also related to the Conflict Theory, identifies how the top-down approach and unequal access to AI create disparities and marginalization, contradicting the Reduced Inequalities objective. Meanwhile, Re-negotiating Pedagogical Authority, which ties to Bourdieu's concept of habitus, concerns teachers modifying their teaching approaches and power in response to technological changes, so that they do not lose their moral values in the digital realm. Everything is connected to Labor, Power, and Sustainability, demonstrating the continuous impact of human actions and institutions on one another and on education. All in all, the model shows that changing education with digital tools is not just about tech; it is a social issue in which empowerment, inequality, and adaptation occur together. To make AI work well in education, we need digital tools, fair working conditions, involvement from everyone, and a focus on humancentered teaching methods to achieve the SDG goals. Discussion As this work's results show, the so-called AI shock in education can be not only a technological advance but also a sociological restructuring of academic work, power, and inequality, especially in the Global South. On the one hand, AI integration was moderately adopted by teachers, which, at the same time, created additional perceptions of job precarity, as defined by Kalleberg (2018), based on which the digital modernization increases but does not reduce professional insecurity. Teachers noted greater anxiety due to opaque AI systems, managerial surveillance, and top-down policy demands, a trend also found in recent scholarship across many countries, including South Korea. (Jung, 2024) and the UK (Smith, et al., 2024), where algorithmic management has confined the distance between empowerment and control. This two-sided effect was also observed in the regression and SEM findings, with Pedagogical Authority (b = 0.37) and Digital Capital (b = 0.20) as very strong positive predictors of teacher performance. In contrast, Job Precarity (b = -0.38) was a quite strong negative predictor. The given dynamic supports the Capital Theory of Bourdieu’s (1986) digital competence has become a symbolic and economic resource that can be transformed into pedagogical legitimacy and professional value. Nevertheless, the inequality in the dissemination of this capital throughout institutional pickings resembles Bourdieu & Passeron's (1990), Social Reproduction Theory, demonstrating that AI-incre integrated might widen systemic gaps between colleges with resources and those who are under-resourced. International studies: studies by China supports this trend (Zhu & Aslan, 2023) and Finland (Vigier & Bryant, 2025) indicate that faculty in elite institutions use AI to be innovative and conduct research. In contrast, teachers in lower-level and peripheral institutions experience automation anxiety and a lack of training, replicating global patterns of educational labor hierarchies. Structural Equation Model (SEM) also demonstrates that the mediating performance of communication and social behavior constructs is compatible with the theory of Social Learning (1977) of Bandura and Giddens' Structuration Theory (1984),) and Structuration Theory. In addition to the features of communication that predict faculty mentorship, time management, and character, these features shape the moral and emotional environment of classrooms, reflecting how human agency and institutional norms constitute digital pedagogies. This result is consistent with international empirical data on Australia (Dianati, 2024) and the U.S. (Yang & Du, 2024) showing that teacher relational and emotional labor is still the core of the maintenance of the quality of education under the mediation of technologies. The interpretation is further elaborated through the data from the qualitative analysis; teachers described AI as a hope-and-fear tool, reflecting the twosided nature of technological change. According to the Hierarchical Tree Model, digital transformation occurs across three interdependent areas: emotional, institutional, and pedagogical. At the very first level, Technological Anxiety and Institutional Coercion were used to explain how policy-based requirements undermine autonomy, thereby validating Tambulasi's (2025) concept that power circulates through discourses of efficiency and accountability. However, at the intermediate layer, Digital Capital Copyright © ISRG Publishers. All rights Reserved. DOI: 10.5281/zenodo.17596577 172 as Empowerment and Re-negotiating Pedagogical Authority demonstrate adaptive agency: educators who learn and engage with AI-based tools to redefine their authority exemplify what Giddens would call reflexive modernization. This adaptive capability is consistent with the results in the OECD countries, where the teacher AI literacy is associated with increased professional satisfaction and the ability to innovate (OECD, 2024). Nonetheless, the bottom segment of the model - including gender imbalance, policy impetus, and moral direction - shows that digital empowerment has not been proportionately distributed. The female teachers, especially those in public-sector colleges, reported uneven workload growth and a lack of institutional support, findings supported by the Global Education Monitoring Report (Arnold and Rahimi, 2024), which addresses the intersection of gender and infrastructure disparities to limit equitable digital transformation. The Cluster Map generalizes these results within a larger sociological model that connects AI integration to the UN Sustainable Development Goals (SDGs 4, 8, and 10). The evidence that is triangulated by the study indicates that although AI can improve Quality Education (SDG 4) by developing skills and creating new pedagogical opportunities, it is also a threat to Decent Work (SDG 8), as it encourages the digital precarity that is normalized, and contributes to Inequalities (SDG 10) by uneven institutional resources. Such compound scheduling is indicative of the discussions concerning globalization, as expressed by Brown (2003), and it poses a threat of further entrenching labor and learning inequalities. The Pakistani context further exacerbates this tension: AI is not adopted following the gradual, bottom-up technological evolution model of the Western system, but is imposed and coercive by the state policy, creating a form of what Zeng (2021) The term coercive isomorphism refers to institutional adherence driven by legitimacy rather than capacity. As a result, educators usually practice AI performatively, thereby creating a disconnect between the symbolic adoption and real empowerment. Similar patterns are also found in comparative international research, like Molosi-France and Makoni (2020) of the case of African universities and Ordóñez de Pablos (2024) The case of Indonesia shows that the AI-driven modernization of education replicates North-South disparities worldwide in the name of innovation. On the whole, the study contributes to the sociological analysis of education by situating AI within labor-power relations, with technology restructuring the professional hierarchy, pedagogical authority, and the division of emotional labor. It contributes to (Preston, 2021), when educational work is spread through data and controlled algorithmically. On the ground, it reveals how the topdown process of digital Pakistan has the potential to reproduce old technological disparities. However, the results also indicate such avenues of resilience: educators who develop digital capital, work together, and re-establish their habitus in this new area demonstrate increased power and achievement. Therefore, the ongoing need for sustainable digital transformation in education requires a sociotechnical balance, as the humanistic pedagogy, the equal policy, and the development of digital skills progress in parallel. In line with SDGs 4, 8, and 10, the paper argues that the adoption of AI could lead not only to greater efficiency but also to greater justice, provided it is grounded in the dignity of labor, capacity building, and ethical governance. By doing so, it is part of an increasingly global discussion that proposes a just transition to digitalisation in education, where human labour is given equal weight with machine learning, and where AI in education remains inherently human. Conclusion The current paper concludes that the implementation of Artificial Intelligence (AI) in higher education in Pakistan is not merely a technological change, but a radical social upheaval of academic labor, authority, and inequality. The evidence demonstrates that, while implementing AI enhances digital capital and pedagogical authority, which, in turn, improves teacher performance, it also leads to job precarity and institutional dependency, particularly in circumstances where resources are scarce. The given paradox is similar to findings from studies conducted across the globe over the last several years in South Korea, China, and the United Kingdom, which stress that AI's educational capabilities are commonly mediated by socioeconomic status. The Pakistani context has shifted the professional identity and agency into the AI shock, and teachers have to redefine their habitus in a sector controlled by digital technology. The paper validates the claim that digital change is a site of empowerment and social reproduction, drawing on Bourdieu's Capital Theory, Kalleberg's Precarity Framework, and Giddens's Structuration Theory. Data from mixedmethods research demonstrate that the idea of a sustainable and fair application of AI necessitates a balance among technological innovation, humanistic values, and structural justice. Lastly, this paper restructures AI as a socio-labor phenomenon that can only be democratized through education, provided digital competence, institutional support, and policy justice are unified within a common arena. Recommendations Based on the findings and in line with SDGs 4 (Quality Education), 8 (Decent Work), and 10 (Reduced Inequalities), the research proposal is a multi-level policy and institutional response to achieve an equitable digital shift in higher education. At the policy level, governments and higher education commissions should also go beyond the formal requirements of digitalization to include capacity-building initiatives that deliver inclusive AI literacy, invest in infrastructure, and ensure equal reallocation of workloads, particularly in marginalized institutions. Universities must collegially build participatory versions of digital governance, in which teachers collaboratively develop AI policies that reduce the demands of AI implementation through coercion and ensure professional autonomy. Ethically, training programs must integrate the ethical application of AI, critical digital pedagogy, and socioemotional resilience, as educators are prepared to operate across the technical and ethical facets of AI-enabled teaching. Furthermore, North and South international institutions should be more collaborative with UNESCO and the OECD to close the digital divide and promote international solidarity in education technology. Lastly, research universities should institutionalize systems of consistent evaluation to analyze the social impacts of AI on labor, gender, and equity, ensuring that digital advances are also human dignityand sustainable development-oriented. These strategies, taken together, require human-centered digitalization, i.e., a digitalization in which teachers are not regarded as passive consumers of AI systems, but as agents of sociotechnical change and social justice in education. References 1. Ahmad, M., Ali, A. N., Sattar, F., & Hussain, H. (2025). National Spatial Data Infrastructure as a Catalyst for Good Governance and Policy Improvements in Pakistan. Copyright © ISRG Publishers. All rights Reserved. DOI: 10.5281/zenodo.17596577 173 ISPRS International Journal of Geo-Information,, 14(9), 324. 2. Ajibade, O., & Ibietan, O. (2016). Public bureaucracy and service delivery in Nigeria: The neo-Weberian explanation. The Public Administration and Social Policies Review,, 2(17), 5-18. 3. Akyeampong, K. (2022). Teaching at the bottom of the pyramid: Teacher education in poor and marginalized communities. . Learning, marginalization, and improving the quality of education in low-income countries, , 77-111. 4. Arnold, B., & Rahimi, M. (2024). The Global Status of Teachers. 5. Baildon, M. (2025). Confronting climate denial in higher education to promote sustainable futures. Journal of Adult and Continuing Education,, 31(1), 7-21. 6. Beech, D. (2021). Deskilling in art and the problem of ideology. The Quarterly of the Institute of Art History at the University of Wrocław,, 59(1), 87-98. 7. Birbirso, D. T. (2013). Technology for Empowering or Subjugating Teachers: Analysis of Ethiopia's Education Reform Discourse Practice. Journal for Critical Education Policy Studies (JCEPS), , 11(4). 8. Bourdieu, P. &. (1990). Reproduction in education, society and culture (Vol. 4). Sage. 9. Bourdieu, P. (1986). The forms of capital. Cultural theory: An anthology. 10. Brown, P. (2003). The opportunity trap: Education and employment in a global economy. European Educational Research Journal, 2(1), 141-179. 11. Castro, A. J. (2022). Teachers of color and precarious work: The inequality of job security. Labor Studies Journal, 47(4), 359-382. 12. Chakravarty, N. (2022). AI Deployment Models for Augmenting Higher Education: Matching Institutional Characteristics and Goals with Innovative Markets. . University of Pennsylvania. 13. Dianati, S. (2024). Indigenising the curriculum: Benchmarking across the Australian higher education sector. In HERDSA Annual Conference Together We Innovate. 14. Durrani, N., & Halai, A. (2018). Dynamics of gender justice, conflict and social cohesion: Analysing educational reforms in Pakistan. International Journal of Educational Development,, 61, 27-39. 15. Entrich, S. R. (2020). Worldwide shadow education and social inequality: Explaining differences in the socioeconomic gap in access to shadow education across 63 societies. International Journal of Comparative Sociology, 61(6), 441-475. 16. Gerrard, J., & Farrell, L. (2014). Remaking the professional teacher: Authority and curriculum reform. Journal of Curriculum Studies,, 46(5), 634-655. 17. Giddens, A. (1984). Elements of the theory of structuration. Practicing history: new directions in historical writing after the linguistic turn. 18. Gordon, C. (1980). Power/knowledge. New York.: Pantheon. 19. Ikpuri, E. O. (2023). The role of social reproduction theory in understanding the issue of inequality in the United States education system. International Journal of Latest Research in Humanities and Social Science,, 6(9), 140-146. 20. Jung, S. (2024). The Datafied School in the Neoliberal Era: Pandemic Shifts in South Korean Education Policy. International Journal of Communication,, 18(20). 21. Kalleberg, A. L. (2018). Job insecurity and well-being in rich democracies. . The Economic and social review, (Autumn), 49(3), 241-258. 22. Kalleberg, A. L., & Vallas, S. P. (2017). Probing precarious work: Theory, research, and politics. In Precarious work. Emerald Publishing Limited. 23. Khalil, I. U., Hena, S., Ghani, U., Ullah, R., Jan, I., Rauf, A., & ... & Jingdong, L. (2021). Development and sustainability of rural economy of Pakistan through local community support for CPEC. Sustainability,, 13(2), 686. 24. Mancaniello, M. R., & Lavanga, F. (2024). Adolescence in the Italian labour market: In search of an equilibrium among instability, uncertainty, and AI challenges. Social Sciences,, 13(12), 688. 25. Merisalo, M., & Makkonen, T. (2022). Bourdieusian ecapital perspective enhancing digital capital discussion in the realm of third level digital divide. Information Technology & People,, 35(8), 231-252. 26. Molosi-France, K., & Makoni, S. (2020). A partnership of un-equals: global south–north research collaborations in higher education institutions. Modern Africa: Politics, History and Society,, 8(2), 9-24. 27. Natsiopoulou, E. (2025). Poisoned schools and automated students: The crisis of social reproduction. . Educational Philosophy and Theory,, 57(5), 410-423. 28. Nawaz, A. (2012). E-Learning experiences of HEIs in advanced states, developing countries and Pakistan. Universal Journal of Education and General Studies,, 1(3), 72-83. 29. OECD. (2024). Integrated care. European Commission. 30. Ordóñez de Pablos, P. (2024). Digital Technologies for a Resource Efficient Economy. Ed.: IGI Global. 31. Pappa, D., Pawlowski, J., Clements, K., & & Sotiriou, S. (2024). TECHNOLOGY-ENHANCED EDUCATION IN SCHOOLS: CHALLENGES TO WELL-BEING. In ICERI2024 Proceedings IATED, 7362-7371. 32. Pitzalis, M., & Porcu, M. (2024). Digital Capital and Cultural Capital in education: Unravelling intersections and distinctions that shape social differentiation. . British Educational Research Journal,, 50(6), 2753-2776. 33. Preston, J. (2021). Artificial intelligence in the capitalist university: Academic labour, commodification, and value. Taylor & Francis., 182. 34. Rane, N. (2024). Enhancing the quality of teaching and learning through Gemini, ChatGPT, and similar generative Artificial Intelligence: Challenges, future prospects, and ethical considerations in education. TESOL and Technology Studies, 5(1), 1-6. 35. Rothstein, R. (2013). Why children from lower socioeconomic classes, on average, have lower academic achievement than middle-class children. Closing the opportunity gap: What America must do to give every child an even chance, , 61-74. 36. Roy, S., & Paul, S. K. (2023). Revolutionizing education: How artificial intelligence is transforming the