The Human Rights Risks of Generative Artificial Intelligence and Legal Governance
The inherent tension between the self-empowering nature of artificial intelligence(AI)and its need for regulation has given rise to a potent force:digital power.Shielded by digital capital and the"black box"nature of algorithms,this power permeates all facets of society,subtly shifting the balance of authority from human to technological control and raising the specter of an"intelligent Leviathan."The advent of generative AI signifies a paradigm shift from task-specific,decision-making AI to a more versatile and general-purpose form.While this evolution unlocks unprecedented creative and autonomous potential,it also disrupts the established trajectory and funda-mental logic of AI development.This shift presents significant challenges in aligning AI with ethical principles,legal frameworks,and societal values.Reforming and inno-vating AI governance strategies is thus paramount.Failure to bring AI development and deployment under the rule of law risks technological alienation,unchecked expan-sion of digital power,and the erosion of fundamental human rights such as equality and freedom.Two primary models currently guide AI governance:ex-ante governance based on risk prevention and decentralized governance based on specific elements and scenarios.The former prioritizes comprehensive planning and risk mitigation,while the latter focuses on precise control within defined application contexts.However,the rapid evolution of AI,characterized by increasing uncertainty and the emergence of generative AI with its generalized capabilities,cross-modal functionality,and emer-gent intelligence,challenges these traditional governance approaches.Element-based,scenario-focused governance and risk prevention-centric strategies may prove insuffi-cient in addressing the unique challenges presented by this new generation of AI.To effectively counter the potential"Leviathanization"of intelligent technologies,criti-cally reassessing and re-calibrating current AI governance frameworks are essential.A human rights-centered approach,carefully considering the current state of generative AI,should guide the development of flexible,adaptive,and holistic institutional designs.Strengthening ethical guidelines,enacting comprehensive legislation,and implementing agile governance mechanisms are crucial steps towards a more nuanced and effective approach to AI governance.These measures will help ensure technologi-cal advancement goes hand in hand with respecting and protecting fundamental human rights.Looking ahead,establishing a robust theoretical framework and institutional system for digital human rights,tailored to the specific context of China,is crucial to prevent the undue expansion of digital power and ensure a future where AI serves humanity.
generative artificial intelligenceChatGPTtechnological powerdigital human rightsagile governance
万方数据
维普
