Survey of 400 Indian corporate workers finds AI adoption alone does not guarantee sustainable employment; firms that combine AI with reskilling, adaptable employees and productivity improvements report stronger employment sustainability.

Main Finding

The paper finds that organizational productivity and employee adaptability are the strongest predictors of employment sustainability in AI-enabled corporate settings. Mere AI adoption is not sufficient to secure sustainable employment outcomes; continuous reskilling, workforce adaptability, and productivity optimization are critical. The authors operationalize these relationships through a new AI Workforce Transformation Index (AIWTI) and recommend human-centered AI governance and reskilling programs for organizations and labor policy.

Key Points

• New index: AI Workforce Transformation Index (AIWTI) constructed via min–max normalization and equal weighting of five dimensions: AI Adoption, Workforce Reskilling, Employee Adaptability, Organizational Productivity, and Employment Sustainability.
• Sample: 400 valid survey responses (employees and managers) from Indian corporate sectors (IT, BFSI, manufacturing, healthcare, retail, telecom, consulting); initial outreach 650, 452 responses, 400 after screening.
• Reliability/validity: Overall Cronbach's alpha = 0.903 (excellent). Measurement validity checked using AVE, Fornell–Larcker, and HTMT criteria; reliability also assessed via composite reliability.
• Main statistical results:
  • Pearson correlations: all constructs positively and significantly correlated (p < 0.001).
  • Multiple regression predicting Employment Sustainability: Organizational Productivity β = 0.361, Employee Adaptability β = 0.294. Model R² = 0.647, F = 92.43, p < 0.001 (predictors explain ~64.7% of variance).
• Hypotheses tested included links from AI adoption to productivity and employment dynamics, mediation by reskilling, and moderation by ethical AI governance; core conclusion emphasizes mediating/conditional role of reskilling and adaptability.
• Policy/managerial recommendations: adopt human-centered AI governance, implement continuous reskilling, use AIWTI to assess organizational readiness and socio-economic workforce outcomes.

Data & Methods

• Research design: Quantitative, explanatory, cross-sectional; deductive approach grounded in Technology–Organization–Environment (TOE) and Human Capital Theory.
• Data collection: Online survey (Google Forms) during Jan–Mar 2026; pilot with 50 respondents; informed consent, anonymized responses, checks to reduce duplication; respondents had ≥1 year organizational experience.
• Sampling: Stratified random sampling across industries and organizational levels.
• Measurement:
  • Constructs used (as reported): AI Adoption, Workforce Reskilling (ERS), Employee Adaptability, Organizational Productivity (Workforce Productivity), Employment Sustainability (EMS). The paper also includes Job Displacement Perception (mediating) and Ethical AI Governance (moderating) in the conceptual model.
  • Items: Likert 1–5. The manuscript contains some internal inconsistencies about item counts (mentions 20 items in one place, a 35-item table elsewhere); overall reporting indicates multi-item scales adapted from prior literature.
• Analytical tools: IBM SPSS Statistics 29 and SmartPLS 4.0. Main techniques: descriptive statistics, reliability (Cronbach’s alpha, composite reliability), convergent/discriminant validity (AVE, Fornell–Larcker, HTMT), PLS-SEM with 5,000 bootstrap resamples, model fit/evaluation reported (R², f², Q², SRMR). Multi-group industry comparisons and sensitivity analysis of predictor contributions are described.
• Index construction: AIWTI computed by min–max normalization of latent-dimension scores followed by equal weighting and aggregation into a single readiness/transformation metric.

Implications for AI Economics

• Complementarity matters: The results reinforce the economic view that AI is complementary to skilled labor and organizational capabilities. Productivity gains from AI translate into sustainable employment primarily when firms invest in reskilling and build employee adaptability.
• Policy levers: Labor-market and industrial policies should prioritize subsidized/reskilling programs, incentives for human-centered AI governance, and mechanisms to monitor firm-level workforce readiness (AIWTI-like metrics) to reduce displacement risks and inequality.
• Measurement & monitoring: AIWTI provides a practical, firm-level composite metric for policymakers and managers to track readiness and socio-economic outcomes of AI adoption. However, index construction choices (equal weighting, min–max normalization) have normative implications—sensitivity testing and alternative weighting schemes should be considered in policy use.
• Distributional concerns: Cross-sector heterogeneity (IT vs manufacturing vs healthcare) implies different labor-market impacts; targeted sectoral strategies will be more effective than one-size-fits-all approaches.
• Research & evaluation: Findings underscore the need for longitudinal and causal studies to assess dynamic effects of AI on wages, employment levels, and inequality. Cross-sectional self-reports limit causal inference; future work should use panel data, administrative employment records, and richer productivity measures.
• Practical economics takeaway: Investment in human capital (reskilling, adaptability) amplifies the economic returns to AI at the firm level and can mitigate negative employment externalities—suggesting a role for coordinated public–private investment in workforce development to capture AI-driven growth inclusively.

Limitations to note (reported or implied): cross-sectional design limits causal claims; reliance on self-reported survey data; some inconsistencies in reported item counts/construct roles in the methods section; sample limited to Indian corporate settings — external validity to other countries/SMEs requires caution.