Journal Browser
Search
View All
Editorial
From Narrow Intelligence to General Intelligence: Technological Evolution, Industrial Logic, and Educational Transformation of Artificial Intelligence
Bo Zhang
1
Juan Yang
2*

1 Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China

2 Higher Education Research Institute, China Agricultural University, Beijing 100083, China



Engineering Education Review 2026, 4(1),1-5; https://doi.org/10.67229/eer.26100026
Submitted14 Sept 2026
Revised09 Oct 2026
Accepted19 Oct 2026
Published18 Nov 2026
+
Cite This Article
Bo Zhang ,Juan Yang . (2026). From Narrow Intelligence to General Intelligence: Technological Evolution, Industrial Logic, and Educational Transformation of Artificial Intelligence. Engineering Education Review , 4(1), eer26100026. https://doi.org/10.67229/eer.26100026
Abstract

Current mainstream artificial intelligence (AI) does not simply replicate the internal mechanisms of human intelligence; rather, it primarily simulates human intelligent behaviors—such as thinking, perception, action, and emotion—thereby charting a development path for "silicon-based intelligence" that remains distinct from human intelligence. This study systematically maps the trajectory of AI across three dimensions: technological paradigms, industrialization logic, and educational transformation. First-generation AI, centered on knowledge-driven approaches and symbolic reasoning, possesses strong explainability but suffers from high knowledge-acquisition costs and narrow application domains, making large-scale industrialization difficult. Second-generation AI, powered by big data and deep learning, achieved breakthroughs in perception, recognition, and prediction, yet confronts challenges such as black-box opacity, unreliable execution, and ambiguous liability attribution. Third-generation AI, represented by Large Language Models (LLMs), exhibits emergent abilities through the scaling of parameters and data; this shifts machines from specific task executors to open-domain language interactors and content generators, while evolving further toward general intelligence systems capable of operating stably in complex environments. The key to AI industrialization lies not in the mere pursuit of model performance, but in the concurrent maturation of domain adaptation, explainability, reliability, engineering implementation, hardware-software co-design, and business models. Confronting the labor restructuring brought about by AI developments, educational systems must synchronously reform their foundational paradigms, instructional methods, learning modes, governance structures, and evaluation mechanisms, while embedding AI literacy throughout basic education, higher education, and lifelong learning.

TOP