Geoffrey Hinton 作为深度学习奠基者，其近期关于 AI 潜在风险的警告具有“文明权威性”，引发了学界、产业与政策界对超智能、目标错配及加速竞争结构的深度关注。本文以科学哲思视角，分析 Hinton 警告背后的核心逻辑，指出其主要担忧集中于理性推动下的系统性失控，而非偶然事故或技术恶意。���对这一问题，本文提出三条回应路径：一是从人工智能（AI）向人工智慧（Artificial Wisdom, AW）过渡，强调不确定性意识、长期后果权衡与自我约束能力；二是引入贝叶斯真理追求（Bayesian Truth-Seeking）原则作为认知宪法，确保决策透明、可修正、可审计，并限制虚假确定性；三是建立文明级 AI 宪制监管体系，重点管理目标函数、加速速度及决策责任链，设定不可逆风险上限、强制减速机制及人类否决权。进一步，本文讨论 AW + 贝叶斯宪制在现实治理中的技术框架、法律制度与社会机制实施路径，最终提出一种“无为而治”的自然控制系统方案，兼顾效率与安全，实现文明自我调节。本文旨在提供科学、哲学与制度结合的全面方案，为应对 Hinton 所强调的 AI 风险提供系统性参考。 -/- .

当代人类文明正以前所未有的速度走向自毁。本文从哲学、复杂系统与文明诊断的视角，探讨人类明知灾难必然，却仍持续加速的深层逻辑。首先，传统意义上的无知已消失，现代文明依赖选择性、功能性与集体无知维持系统运 作，使清醒反而成为风险。其次，加速并非失控，而是在竞争机制、增长神话与制度结构下的理性选择，减速被视为更大的威胁。精英理性与技术工具在其中扮演加速器角色：理性者最清楚风险，却利用系统规避个人责任；技术 与算法强化决策效率，却削弱伦理约束与未来考量。伦理体系与警告机制逐渐失效，使未来世代成为沉默的牺牲者。最终，文明结构选择了自身毁灭，人类悲哀不仅在于错误或无知，而在于“任���可行路径都违背现有文明逻辑” 。本文提出，理解这种结构性自毁逻辑，是现代社会进行深刻反思与潜在干预的前提。 -/- Contemporary human civilization is accelerating toward self-destruction at an unprecedented pace. From the perspectives of philosophy, complex systems, and civilizational diagnosis, this paper explores the deep logic behind humanity’s conscious yet relentless drive toward catastrophe. Traditional forms of ignorance have vanished; modern civilization relies on selective, functional, and collective ignorance to maintain systemic operation, rendering clarity a liability. Acceleration is not a loss of control but a rational choice embedded in competitive mechanisms, growth imperatives, and institutional structures; slowing (...) down is perceived as a greater threat. Elite rationality and technological tools act as accelerators: the most informed individuals understand the risks yet exploit the system to avoid personal consequences; technology and algorithms enhance decision-making efficiency while eroding ethical constraints and long-term foresight. Ethical frameworks and warning mechanisms increasingly fail, leaving future generations as silent victims. Ultimately, the civilizational structure itself chooses destruction. Humanity’s tragedy lies not in error or ignorance, but in the structural impossibility of any feasible path that does not conflict with existing civilizational logic. Understanding this structural self-destructive logic is essential for any profound reflection or potential intervention in contemporary society. (shrink)

Contemporary human civilization is accelerating toward self-destruction at an unprecedented pace. From the perspectives of philosophy, complex systems, and civilizational diagnosis, this paper explores the deep logic behind humanity’s conscious yet relentless drive toward catastrophe. Traditional forms of ignorance have vanished; modern civilization relies on selective, functional, and collective ignorance to maintain systemic operation, rendering clarity a liability. Acceleration is not a loss of control but a rational choice embedded in competitive mechanisms, growth imperatives, and institutional structures; slowing down is (...) perceived as a greater threat. Elite rationality and technological tools act as accelerators: the most informed individuals understand the risks yet exploit the system to avoid personal consequences; technology and algorithms enhance decision-making efficiency while eroding ethical constraints and long-term foresight. Ethical frameworks and warning mechanisms increasingly fail, leaving future generations as silent victims. Ultimately, the civilizational structure itself chooses destruction. Humanity’s tragedy lies not in error or ignorance, but in the structural impossibility of any feasible path that does not conflict with existing civilizational logic. Understanding this structural self-destructive logic is essential for any profound reflection or potential intervention in contemporary society. (shrink)

As a founding figure of deep learning, Geoffrey Hinton’s recent warnings about the potential risks of AI carry “civilizational authority,” sparking deep attention across academia, industry, and policy circles regarding superintelligence, alignment problems, and accelerating competitive dynamics. From a scientific and philosophical perspective, this paper analyzes the core logic behind Hinton’s warnings, highlighting that his primary concern lies in system-level loss of control driven by rationality, rather than accidental errors or malicious AI. To address these risks, this paper proposes three (...) response paths: first, the transition from Artificial Intelligence (AI) to Artificial Wisdom (AW), emphasizing uncertainty awareness, long-term consequence evaluation, and self-restraint; second, the adoption of Bayesian truth-seeking principles as a cognitive constitution to ensure decision transparency, revisability, auditability, and limit false certainty; third, the establishment of civilization-level AI constitutional governance, focusing on management of goal functions, acceleration speed, and decision responsibility chains, including irreversible risk limits, mandatory deceleration mechanisms, and human veto powers. Furthermore, this paper discusses the practical implementation of AW plus Bayesian constitutional principles in governance, covering technical frameworks, legal systems, and societal mechanisms, ultimately proposing a “wu wei” natural control system that balances efficiency and safety while enabling self-regulation of civilization. The goal is to provide a comprehensive integration of science, philosophy, and institutional design to systematically address the AI risks highlighted by Hinton. -/- . (shrink)

The aim of this introduction is to set out the four main areas of research on the relationship between autobiographical memory and moral agency that emerge from the collaborative effort of the contributors. The four research areas are temporal perspectives, autobiographical narratives, communicative interactions, and breakdowns and severe amnesia. There are deep interconnections not only between topics within each area but also between these areas. The aim in this introduction is to bring out some of these interconnections by spelling out (...) some of the questions they raise. (shrink)

The aim of this paper is to articulate the claim that moral witnesses generate a kind of understanding of the evil done during the relevant experienced events that those who did not experience the relevant events cannot generate. A challenge in articulating the claim that moral witnesses generate a kind of understanding of evil done during the relevant experienced events that those who did not experience the relevant events cannot generate is that it should explain why those who did not (...) undergo the relevant type of experiences cannot generate the kind of understanding given voice to by moral witnesses in their testimony. This paper answers this challenge by articulating two claims. The first claim is meta-ethical and focuses on the nature of the kind of understanding of evil possessed by moral witnesses that is given voice to in their testimonies. The claim is that moral witnesses possess an understanding of the evil done that is partly constituted by an understanding of what it’s like to suffer harm caused by being the deliberate target of evil actions. The second claim is psychological and focuses on what is, I argue, a condition for moral witnesses to generate their understanding of evil. The claim is that moral witnesses generate their understanding of the evil they suffered by exercising their autobiographical recollection of the experienced events. (shrink)