摘要： 随着生成式人工智能在社会治理、公共决策与日常生活场景中的广泛应用，其对生活世界的介入方式也正在发生转变，即由对既有事实的技术性处理，转向基于概率模型与算法推理的前瞻性预测。在可能世界理论与模态逻辑的分析框架下，生成式人工智能可以通过对经验事实的抽象建模与概率演算，将原本依赖主体经验、价值判断与规范理解的生活世界，重构为一组可计算、可比较、可预期的模态化可能世界。这种转变的价值在于提升了决策效率与风险控制能力，但转变的可能风险是对规范基础构成深刻挑战。生成式人工智能预测功能的滥用、全息认知的误导以及模态演算的信任风险，是短时间内学界无法回避的议题。对此，需要从生成式人工智能参与生活世界模态化边界范围的廓清、生成式人工智能预测模型设计与更新道德审查的强化，以及生成式人工智能自我迭代与自主运行法律约束的完善等层面入手，构建生活世界模态化的规范话语。

关键词:

Abstract: With the widespread application of artificial intelligence in social governance, public decision-making, and daily life scenarios, its mode of intervention in the lifeworld is undergoing a shift: From the technical processing of existing facts to prospective prediction based on probability models and algorithmic inference. Within the analytical framework of possible worlds theory and modal logic, generative artificial intelligence can reconstruct the lifeworld—which originally relied on subjective experience, value judgment, and normative understanding—into a set of computable, comparable, and anticipatable modalized possible worlds. This is achieved through the abstract modeling of empirical facts and probabilistic calculus. The value of this transformation lies in enhancing decision making efficiency and risk control capabilities, but its potential risks pose a profound challenge to normative foundations. The imminent issues are the misuse of the predictive functions of generative artificial intelligence, the misguidance of holistic cognition, and the trust risks associated with modal calculus. These are unavoidable in the short term. Addressing them requires starting from clarifying the boundaries of the lifeworld’s modalization by generative artificial intelligence, strengthening ethical review in the design and updating of generative artificial intelligence prediction models, and improving legal constraints on the self-iteration and autonomous operation of generative artificial intelligence, thereby constructing a normative discourse for the modalization of the lifeworld.

Key words: