Lu Xiangfeng: The Ethical Dimension of Artificial Intelligence Governance in the New Era
General Secretary Xi Jinping, while presiding over the twentieth collective study session of the Political Bureau of the CPC Central Committee, emphasized the need to "comprehensively promote artificial intelligence (AI) technological innovation, industrial development, and empowering applications; improve AI regulatory systems and mechanisms; and firmly grasp the initiative in AI development and governance." In the era of AI, issues of AI governance have transcended technical boundaries to become a core agenda of global governance affecting the future development of humanity. Since the inception of AI, its manufacture, application, and status as a disruptive technology have been accompanied by ethical concerns, causing traditional frameworks of moral governance to face a series of new challenges. Amidst the reality of AI empowering all sectors and industries, a comprehensive and forward-looking demand for ethical governance has emerged. It can be said that the construction of AI ethics and its governance pathways is not merely a phased task of technical governance, but an inherent requirement for defining the ethical boundaries of the future digital-intelligent society.
Constructing Ethical Norms for AI Governance
The core objective of constructing AI ethical norms is to embed moral judgments and codes of conduct into intelligent systems that align with the laws of human social development, ensuring that they consistently follow the fundamental principle of safeguarding the collective interests of humanity throughout the entire process of empowering economic and social development. In essence, among the many fields of applied ethics, none is as closely linked to the concept of responsibility as the ethics of science and technology. To realize this vision, the field of ethics has undergone a theoretical evolution—from Asimov's "Three Laws of Robotics" to Wiener's "machines should serve human needs," and further to the "value alignment" advocated by thinkers like Millett. This research not only identifies the difficulties of human-machine value alignment but also highlights the critical role of moving from "value alignment" toward "value symbiosis." However, due to the spatiotemporal limitations of human cognition, inherent difficulties in value encoding, and the technical boundaries of algorithmic operation, AI ethical theory faces significant operational constraints. Undoubtedly, the construction of AI ethical norms cannot stop at making machines align with static human values; rather, it should move forward-looking toward a new stage of collaborative development where humans and machines achieve mutual understanding, adjustment, and co-evolution through interaction.
The core of the problem lies in how we explore a set of embeddable, verifiable, and sustainable AI ethical norms against the historical backdrop of technical complexity, algorithmic opacity, and diverse value systems. This requires relevant responsible subjects to transcend pure technical boundaries and build a systemic project that requires technical breakthroughs, philosophical reflection, institutional innovation, and social dialogue to move forward in coordination.
Since 2024, the intensive release of global documents such as the Beijing AI Safety International Consensus, the Shanghai Declaration on Global AI Governance, and the Tianjin Declaration of the Council of Heads of State of Member States of the Shanghai Cooperation Organization [1] marks the formation of a key consensus in the international community regarding AI ethical identity. These aim to jointly construct an ethical governance framework for the AI era centered on the protection of human rights, social justice, and personal dignity. However, the implementation of this consensus faces deep challenges. Traditional concepts of "value alignment" are deeply rooted, while real-world value synergy is actually a dynamic, two-way process of adjustment and symbiosis. Given this, the metaphysical construction of AI ethical norms must trace the historical questions of how AI ethics took shape, respond to how they establish their foundation, resolve how they overcome difficulties, and structure how they adapt to change. These inquiries constitute an internal "value compass" for AI, making it possible for AI to truly become a trustworthy partner of civilization and thereby build a future of human-machine prosperity where intelligence serves the common good.
In the practical construction of AI ethics, three possibilities exist: top-down normative implementation, bottom-up empirical learning, and a hybrid approach merging the two. These three possibilities not only shape the developmental direction of AI ethical theory but also profoundly influence AI ethical practices worldwide. Top-down implementation involves encoding existing human ethical principles into computational symbols and embedding them into AI systems through algorithmic instructions to regulate the behavior of AI agents. Bottom-up empirical learning involves simulating evolution and moral acquisition, using machine learning and system-level organizational evolution to allow AI to induce moral principles from specific contexts and form decision-making capabilities, ultimately completing an autonomous evolution from experience to norms. The hybrid approach integrates both methods. Based on the guidance of ethical theories, it dynamically generates and optimizes ethical codes through simulation, training, and trial-and-error, while incorporating emotional and cognitive functions to enhance moral decision-making, emphasizing collaborative research across disciplines such as philosophy and computer science.
The Ethical Essence of AI Governance
The burgeoning digital-intelligent era has ensured that AI and its ethical issues are no longer merely metaphysical philosophical speculations but increasingly urgent problems of applied ethics. As a disruptive technology, AI inevitably generates complex ethical disputes and governance dilemmas during its technical evolution and social application.
Looking back at the mainstream theoretical paths and governance practices in AI ethical construction, it is not difficult to identify three interrelated essential problems that persist throughout: First, the construction of AI ethics has blurred the boundary between two fundamentally different behavioral motives: "altruism starting from self-interest" and "pure altruism." In the era of AI, informational capitalism has intensified data monopolies. A few subjects with informational advantages abuse their power, evading responsibility under conditions of unequal rights and responsibilities, and transferring technical risks and catastrophic consequences to the public. This ultimately creates systemic risks where a few profit while the majority bear the costs. Second, AI ethical construction has failed to guide the dynamic flow of cultural concepts during technical evolution and has not transformed the highest spiritual values of human civilization into internal principles that machines can understand and follow. Finally, the process of AI ethical construction has failed to fully identify and respond to the fundamental and urgent threats facing human existence, falling into collective unconsciousness and historical nihilism [2]. In other terms, the dissolution of human spiritual subjectivity by AI will lead to the blurring of the meaning of existence and the hollowing out of a sense of value, causing individuals to lose their exploration of the meaning of life.
Existing research, when responding to AI ethical risks, intentionally or unintentionally avoids two fundamental questions: On one hand, as AI fully integrates into human production and life, it is regulated both by the internal logic of scientific exploration and by the external shaping of capital valorization [3] and social power. How to handle this contradiction—which is both an opposition and a unity—is an essential problem that AI ethical construction cannot avoid. Whether the fundamental driving force of technical development stems from scientific autonomy or is subject to the logic of capital and power directly determines the starting point of AI ethical construction. Structural "labor replacement" and "capacity deprivation" brought by scientific and technological progress have reshaped traditional labor values and the foundations of existence. The technical alienation [4] of the subjectivity of intelligent design makes the monopoly of a few interest groups possible, while the interests of the majority fall into functional disability due to technical exclusion. As the public becomes increasingly dependent on intelligent systems, the cognitive gap continues to dissolve human autonomy and dignity. On the other hand, whether we can construct AI ethical codes when human society falls into value dilemmas and moral challenges due to the over-application of AI is another essential problem awaiting resolution. In short, have ethical rules in the technical field alienated into a form of human-machine "game rules"? If AI ethical construction remains only at the level of technical risk management, ignoring the profound regulation of capital logic and structural inequality behind it, governance paradoxes such as "using algorithms to prevent cheating" will continue to reappear.
Broadly speaking, the process of resolving the difficulties of AI ethical construction is also a process of examining human "ontology" and "co-existentialism," and even more so a deep awareness of and active response to the human spiritual world and existential dilemmas. Effective AI ethical construction must balance a critical examination of technical tools with the continuous renovation of human value practices. Only when AI agents manifest altruistic behavior in their ethical design can the reflexive process of generative AI governance truly begin.
Ethical Innovation in AI Governance
The process of AI ethical construction is essentially a process in which humanity uses moral reason to actively shape the ethical direction of scientific and technological evolution. Specifically, resolving AI ethical dilemmas requires corresponding innovative measures. These can include advocating the principle of risk-sharing to respond to ethical risks and forming an "endosymbiotic" structure. Simultaneously, we must uphold a "co-existentialist" perspective to respond to the crisis of human existence, thereby realizing a shift in AI ethical construction from passive adaptation to active co-construction.
First, we must advocate the "risk-sharing" principle to address ethical risks. Currently, the application of AI technology only synthesizes, simulates, and transforms specific functions of the human brain. At this stage, robot designers and manufacturers are essentially the primary ethical agents; they must jointly bear human-machine risks—such as information asymmetry, power-responsibility asymmetry, and cognitive asymmetry—to avoid the phenomenon of risk transfer. Second, we must form an "endosymbiotic" structure to guide memetic mutations. When dealing with the challenges of memetic mutation, humans and AI technology must construct a closely interdependent symbiotic system. In this system, the two are not in a simple subject-object relationship but in a collaborative relationship of continuous interaction, jointly constituting a dynamically adjusted "endosymbiotic" ecosystem. This symbiotic relationship can not only change people's lifestyles but also help promote the progress of human society. Finally, we must uphold a "co-existentialist" perspective to respond to the crisis of human existence. In the process of AI ethical construction, the coexistence of responsibility and risk must be taken as the highest value goal, systematically preventing and responding to those extreme and existential risks that may endanger the survival of humanity as a whole. In essence, to share moral responsibility, AI behavioral subjects—ranging from nations and ethnic groups to individual designers and manufacturers—must, in some sense, control their behavior and consider their moral responsibility as a whole.
In an era where AI receives high attention while generating significant controversy, we must calmly reflect on the ethical issues in AI governance, rationally plan the development direction of AI, and place the "bridle" of ethical rules upon it. With the profound transformation of AI technology, human modes of existence and moral practices are undergoing a shift from stable subjectivity to fluid symbiosis. The traditional status of humanity as the sole center of cognition and action is gradually dissolving; individual autonomy and dignity face profound challenges, and the foundations of social equity and justice are being eroded. Faced with the systemic ethical risks brought by AI, there is an urgent need for an "emergency ethics" capable of shaping value coordinates, forging a moral bedrock, and constructing an ethical order. In the future society—whether through human-machine collaboration or symbiosis—we need to manifest the brilliance of humanity, gain profound insight into the direction of human development, actively prevent difficulties in social development, and guide the development of the intelligent society onto a healthier and more civilized track. Thus, by resolving the difficulties of AI ethical construction and governance and cultivating a deep sense of crisis and value resilience, we can build a dynamically stable and trustworthy ethical baseline for a future of human-machine symbiosis.
Whether viewed from the trend of AI leading a paradigm shift in scientific research or from the requirements of modernizing the state governance system and governance capacity, AI governance must be considered within the horizon of ethics. Through the effective elucidation of AI ethical construction, we can avoid stopping at the mere alignment of machines with static human values. Through the analysis and grasp of the ethical essence and boundaries of AI, we can maintain a clear-headed attitude toward AI and its governance. Based on the developmental reality of AI, we must continuously enrich and innovate ethical theories to understand, accommodate, and guide AI development, ensuring that the development of the discipline of ethics keeps pace with technological innovation. Only by organically combining ethical construction, ethical governance, and ethical innovation can we promote the upward and "good-oriented" development of AI, creating a win-win situation for the healthy development of technology and the sustained progress of society.
(Author: Lu Xiangfeng, Professor at the School of Marxism, Tianjin Medical University) Source: Guangming Daily (March 23, 2026) Editor: Hui Hui