关键词: 智能体, 价值对齐, 技术规制, 伦理适配, 法律治理

Abstract:

With the implementation of the "Artificial Intelligence+" strategy, ChatGPT and DeepSeek intelligent agents will accelerate their commercialization and provide strong impetus for the development of new quality productivity. Intelligent agents have achieved technological transitions from single task processing to multimodal perception, from rule driven to autonomous learning, and from static response to dynamic programming, which has also led to risks of perception loss, decision-making disorder, and execution disorder. The alignment of values is related to whether intelligent agents can truly serve humanity, and has certain legitimacy, necessity, and feasibility. However, intelligent agents face the practical dilemma of "how to align" and "which values to align with", which is due to the lack of a sound value alignment system. In view of this, it is urgent to adopt a collaborative paradigm of technical regulation, ethical adaptation, and legal governance, that is, to construct a dynamic value recognition system, a systematic value adjustment mechanism, and a flexible value regulation program; Adhering to the ethical mission of putting people first, formulating scientifically reasonable ethical principles, and designing scenario based ethical rules; Determine the standard of value alignment, clarify the nature of value alignment, improve the evaluation method of value alignment, and accelerate the formation of new quality productivity to promote Chinese path to modernization.

Key words: intelligent agents, value alignment, technical regulations, ethical adaptation, legal governance