The Evolution of AI Technological System: Generative, Agentic, and Quantum Paradigms

This study compares the trajectories of three artificial intelligence paradigms: generative, agentic, and quantum AI. A scientometric and technometric analysis applies linear regression and Shannon entropy to publication data and keyword-frequency corpora to measure growth, thematic breadth, and structural diversity. Empirical results show that agentic AI has the most rapid acceleration (B = 3.201, p < 0.05), despite emerging only in 2023. Quantum AI exhibits the most consistent and statistically significant growth (B = 0.70, p < 0.001), driven by advances in hardware foundations, security, and post-quantum applications. Generative AI displays a strong positive trend (B = 1.44, p < 0.001). These trajectories indicate “stack thickening” in the AI system, shifting from passive content generation toward autonomous orchestration, grounding, and multi-agent coordination. Technology policy should promote AgentOps standards, accelerate post-quantum cryptography, strengthen institutional AI literacy, support interoperable infrastructures, and establish adaptive regulation for secure, accountable, reliable, and responsible AI deployment.