The Impact of Artificial Intelligence (AI) in
workplace dynamics
AI in work: A partner or a job threat?
Gokul E & Mesha A
Post graduate students, Department of Statistics, Madras Christian College.
Agenda
⚬ Introduction
⚬ Objective of the study
⚬ Literature Review
⚬ Research Methodology
⚬ Analysis
⚬ Output
⚬ Inference
⚬ Conclusion
⚬ References
INTRODUCTION
⚬ Artificial Intelligence (AI) is no longer the future—it is the present. Organizations
across the globe are increasingly integrating AI into their daily operations,
reshaping industries and transforming job roles.
⚬ However, with this technological progress comes a growing concern: the fear that
AI may eventually replace human workers, leading to job displacement and
insecurity.
⚬ Understanding these dynamics is crucial for governments, companies, and
industries to develop policies that ensure technological advancements lead to
economic growth while safeguarding job opportunities, ensuring that both
technology and human can thrive together.
OBJECTIVE
⚬ The primary goal of this study is to identify the key aspects of Artificial
Intelligence in workplace across different industries, mainly focusing on
corporate and technology sectors.
⚬ To investigate whether the factors identified, impact the Job Security of
employees in the workplace.
LITERATURE REVIEW
⚬ Md. Ashraful Islam, Syed Rayed Ithmam, & Md. Tanvirul Islam"The Impact of Artificial
Intelligence on Daily Life and the Workplace, explores the transformative effects of AI on
various aspects of life and work, emphasizing education, healthcare, and job markets. The
authors argue that while AI promotes efficiency and innovation, it also poses challenges
such as skill gaps and ethical dilemmas. The research uses ANOVA to underscores the
necessity for good governance and policies. Overall, it highlights the profound significance
of AI in shaping future societal norms and employment landscapes.
⚬ Zhiqing Bian examines the significant impact of Artificial Intelligence (AI) on the labor
market, detailing its effects through the displacement and productivity.AI has led to job
losses, particularly in the tech sector. The analysis emphasizes the dual role of AI—while it
displaces traditional jobs, it also creates new industries and roles focused on AI
development and management. The paper discusses challenges such as deepfakes,
biases, and necessary legal measures to address these concerns..
⚬ The study by Necula et al. investigates the impact of (AI) tools on employee productivity
through a survey of 233 employees using advanced data analysis methods including
logistic regression, Random Forest, and XGBoost. Key findings indicate that higher
usage and integration of AI tools significantly boost productivity, particularly among
younger workers. The analysis reveals strong interdependencies between AI tool usage,
innovation, and employee characteristics, with strategic AI adoption being crucial for
maximizing economic benefits. The research suggests that productivity gains do not
necessarily lead to job losses. The authors recommend comprehensive AI integration
supported by targeted training and ethical guidelines to harness AI's full potential.
⚬ Sabina-Cristiana Necula, Doina Fotache, and Emanuel Rieder, published in the journal
Electronics (Volume 13, Issue 3758)”,Assessing the Impact of Artificial Intelligence
Tools on Employee Productivity”. It explores the effects of AI tools on employee
productivity and analyzed how AI tools influence productivity. The conclusion
highlights the significant impact of AI on enhancing employee efficiency and output to
foster interdisciplinary skills in the context of AI, ultimately benefiting economic
growth.
RESEARCH METHODOLOGY
Research Design : The study follows a quantitative approach using Factor Analysis to
identify key dimensions related to AI impact and Structural Equation Modeling (SEM) to
examine relationships between AI-related factors and job security.
Target Population : Employees in corporate and Technology sectors
Sample Size: 423
Data Collection : Survey study through Google Forms. Responses gathered using a
structured questionnaires ( AI related likert scale questions) conducted among
employees focusing on the corporate and technology sectors using Convenience
sampling.
ANALYSIS
⚬ Reliability analysis was conducted to ensure the consistency and accuracy of the
data. This helped confirm that the survey items reliably measured the targeted
aspects.
⚬ Exploratory Factor Analysis (EFA) was conducted to identify underlying patterns in
the data and group related variables into meaningful factors. This helped in
reducing dimensionality and determining the key factors that influence AI in
workplace.
⚬ Confirmatory Factor Analysis (CFA) was conducted to validate the factor structure
identified in EFA. This analysis ensured that the survey items accurately measured
the key influences, confirming the reliability and fit of the model.
⚬ Structural Equation Modeling (SEM) was used to examine how the identified factors
influence the job security of employees.
RELIABILITY ANALYSIS
Reliability analysis was conducted to ensure the
internal consistency and accuracy of the data. The
Scale Reliability Statistics
Cronbach’s Alpha (0.835) confirms that the survey
items are reliable making the results trustworthy for
further analysis.
Cronbach's α
scale 0.835
EXPLORATORY FACTOR ANALYSIS
⚬ ASSUMPTIONS CHECK -
⚬ For Bartlett’s Test of Sphericity p < 0.001, the test result indicates that the data are indeed
suitable for factor analysi..The test is statistically significant.( p should be < 0.05)
⚬ KMO = 0.914: This value indicates excellent sampling adequacy (values closer to 1 are better,
and a value above 0.9 is considered excellent). This suggests that the data are appropriate for
factor analysis.
scree plot
factor Loadings
By princiapal axis extraction method, and varimax orthogonanal rotation, four factors have
been obtained that together explains 87.3% of the total variance. This means the factors
effectively capture most of the variability in the data.
The 4 factors are :
⚬ AI Trust (AT)
⚬ AI Adoption (AA)
⚬ Job Security (JS)
⚬ AI skills and training (ST)
These factors help in understanding how AI influences the workplace,.
CONFIRMATORY FACTOR ANALYIS
⚬ With p-values less than 0.001 for all items,
the factor loadings indicate that all the
variables significantly contribute to their
respective factors (AI Trust, AI Adoption,
Job Security, and AI Skills & Training),This
suggests strong relationships between the
observed variables and the latent factors
they represent.
STRUCTURAL EQUATION MODELING (SEM)
Assumptions :
Fit Measures for Model Evaluation :
⚬ An RMSEA value below 0.05 indicates excellent model
fit. Thus 0.0220 is a good fit.
⚬ CFI (Comparative Fit Index):A CFI value closer to 1
suggests a very good model fit, and 0.997 is excellent.
⚬ TLI (Tucker-Lewis Index) 0.997. A TLI value close to 1 is
ideal for model
These fit indices suggest that the model has a good fit to
the data.
Measurement Model :
⚬ The measurement model
confirms that each observed
variable strongly contributes
to its respective factor,
supporting construct validity.
The five latent constructs (AT,
AA, JS, ST) are
interconnected, highlighting
the high influence of AI in
workplace.
⚬ AI Trust (AT)
⚬ AI Adoption (AA)
⚬ Job Security (JS)
⚬ AI skills and training (ST)
STRUCTURAL EQUATION MODELING (SEM) .
We fix Job security (JS) as the Endogenous variable and then find whether
the other 3 factors influence Job security
HYPOTHESES
• H1: AI trust(AT) positively influences job security (JS)
• H2: AI adoption (AA) positively influences job security (JS)
• H3: AI skills & training (ST) positively influences job security (JS)
⚬ The User Model shows excellent fit with indices like CFI (0.997), TLI (0.997), and IFI (0.997) above
0.99, indicating strong model performance. The NFI (0.982) and RFI (0.979) also reflect good fit.
Although the PNFI (0.848) is lower, it remains acceptable, showing a good balance between fit
and complexity.
⚬ Factor Loadings – All items have high standardized
loadings (β values), ranging from 0.916 to 0.959,
indicating strong associations with their respective
latent constructs.
⚬ Statistical Significance – All factor loadings are
highly significant (p < 0.001), confirming that each
observed variable contributes meaningfully to its
factor.
Thus, this validated measurement model provides a
solid foundation for further structural analysis in the
SEM framework.
AI Trust (AT) AI Adoption (AA) Job Security (JS) AI skills and training (ST)
⚬ The p-values for each relationship are all greater than 0.05, indicating no strong evidence to
support these pathways in the model. All three relationships (AI Trust → Job Security, AI
Adoption → Job Security, AI Skills & Training → Job Security) are weak and statistically non-
significant.
⚬ AI Trust, AI Adoption, and Skills & Training do not directly impact job security (statistically
insignificant).
KEY FINDINGS FROM SEM :
⚬ The SEM model tested the relationships between AI Trust, AI Adoption, AI Skills &
Training, and Job Security.
⚬ The model demonstrated excellent fit to the data, as indicated by fit indices such as
RMSEA (0.0220), CFI (0.997), and TLI (0.997), which confirmed the validity and
reliability of the model.
⚬ AI Trust and AI Adoption did not have a significant impact on Job Security (non-
significant results with p-values > 0.05).AI Skills & Training had a positive, though
non-significant relationship with Job Security, indicating a potential trend but no
strong statistical evidence. With this particular model and sample, there is
insufficient evidence to conclude that AI Trust, AI Adoption, or AI Skills & Training
have a meaningful or statistically significant impact on Job Security.
OVERALL INFERENCE
⚬ Obtained 4 factors : AI Trust, AI Adoption, AI Skills & Training, and Job Security as
the key aspects of Artificial Intelligence in workplace.
⚬ The SEM model assessed the relationships between AI Trust, AI Adoption, AI Skills
& Training, and Job Security within the workplace and demonstrated a very good
fit, with CFI (0.997) and TLI (0.997), indicating that the model structure adequately
represented the data.
⚬ Despite the good model fit, the results indicated no significant impact or no strong
impact of AI Trust, AI Adoption, and AI Skills & Training on Job Security
(all p-values > 0.05).
CONCLUSION
⚬ Through Exploratory Factor Analysis (EFA), we identified four key factors: AI Trust
(AT), AI Adoption (AA), AI Skills & Training (ST) and Job Security (JS) as the key
aspects of Artificial Intelligence in workplace.
⚬ Despite the strong model fit, the relationships between AI Trust, AI Adoption, and AI
Skills & Training with Job Security were found to be non-significant (all p-values >
0.05).This indicates that, in the context of this study, AI Trust, AI Adoption, and AI
Skills & Training do not have a significant direct impact on Job Security.
⚬ The findings suggest that while AI-related factors are important, they may not be the
primary determinants of Job Security in the workplace. Other variables, possibly
external to this model, may need to be considered for a more comprehensive
understanding of Job Security in the age of AI.
Future Directions:
⚬ Future research could explore additional factors or examine how the context of
industries or workplace environment influence the relationship between AI and Job
Security.
⚬ Further research is needed to explore mediating factors influencing job security (e.g.,
company culture, automation policies).
REFERENCE
https://hbr.org/2024/11/research-how-gen-ai-is-already-impacting-the-labor-market
Artificial Intelligence and Robotics: Impact & Open issues of automation in
Workplace,Kanchi Rusia; Shubhyansh Rai; Amrita Rai; Shylaja Vinay Kumar
Karatangi
Unveiling the Future: Exploring the Impacts and Challenges of Artificial
Intelligence in Workplace Applications
Economics of ChatGPT: a labor market view on the occupational impact of
artificial intelligence,Ali Zarifhonarvar ,Journal of Electronic Business & Digital
Economics.
A Research Paper on Impact of AI on Employability in India,Dipak B. Kadve
Jayawant Shikshan Prasarak Mandal
THANK YOU
