The rapid development of generative artificial intelligence is profoundly reshaping the global labor market, compelling higher education institutions to undertake structural reforms. This study constructs a data-driven comprehensive evaluation and optimization framework aimed at precisely quantifying the impact of AI on different occupations and providing scientifically grounded educational strategies. The study first integrates the Nash equilibrium theory with the Analytic Hierarchy Process (AHP) and the Entropy Weighting Method to construct an occupational resilience evaluation model comprising five criterion levels and 13 indicators. The evaluation results show that musicians exhibit the highest resilience due to high barriers to imitation, while software engineers are the most vulnerable because their core skills are highly exposed to large language models. Subsequently, the study utilizes the Bass diffusion model to predict the macro-level penetration rate of Gen-AI and couples it with an improved Lotka-Volterra predator-prey framework to simulate the evolutionary trajectory of the human-machine labor market. Projections indicate that by 2040, employment for software engineers and welders will experience positive growth, whereas musicians will face the severe challenge of niche contraction. Finally, this study developed an educational resource allocation optimization model. Solving it via a sequential quadratic programming algorithm revealed that a collaboration-driven curriculum strategy is the global optimal solution for enhancing graduates’ market competitiveness. Sensitivity analysis confirms that this strategy maintains extremely high robustness under fluctuations in multiple parameters.

generative artificial intelligence, occupational resilience, lotka-volterra model