摘要
人工智能技术被视为改变国际秩序的颠覆性技术,美国政府高度重视人工智能技术的发展。从奥巴马政府到拜登政府人工智能安全议程的演变可以看出,美国人工智能安全议程围绕公共安全、大国竞争与军事战略展开,人工智能安全议程由单一走向多元是美国政府统筹多种安全诉求的结果,反映出美国在人工智能领域推动竞争、保障战略稳定以及试图为合作留有余地的多重战略意图。但安全议程的多元化切分只能作为临时策略,并不能长期保护大国战略稳定不受极端安全议程的影响。当前美国多元安全议程的核心是赢得大国竞争维护技术霸权,大国战略稳定与国际安全合作议程本质上是这一战略目标的“附属品”。安全议程上的本末倒置很容易造成安全治理效率降低、增加政治操弄风险对战略稳定的影响,以及削弱国际社会战略信任与安全共识,阻碍全球层面人工智能安全治理合作的有序开展。面对这一问题,中国在人工智能安全治理领域需要形成自身的安全议程,提出更具公共性的安全主张,在保护自身利益的同时,引导人工智能全球安全治理回归合作共赢的道路。
Artificial intelligence(Al)technology is seen as a disruptive technology that could change the international order,and the U.S.government attaches great importance to its development.The evolution of the U.S.AI security agenda from the Obama administration to the Biden administration shows that it has been revolving around public safety,great power competition,and military stability.The shift from a singular to a multidimensional AI security agenda results from the U.S.government balancing multiple security demands,reflecting its strategic intent to promote competition while ensuring strategic stability and leaving room for cooperation in the AI domain.However,the multidimensional fragmentation of the security agenda can only serve as a temporary strategy.It cannot maintain great power strategic stability from the influence of extreme security agendas in the long run.The core of the current U.S.multidimensional security agenda is to win the great power competition and to retain technological hegemony,with the agendas of great power strategic stability and international security cooperation essentially being subordinate to this strategic goal.Putting the cart before the horse in the security agenda can easily lead to reduced security governance efficiency,increased risks of political manipulation affecting strategic stability,weakened strategic trust and security consensus among the international community,and hindered orderly development of global Al security governance cooperation.Faced with this issue,China needs to form its security agenda in AI security governance and put forward more public-oriented security propositions.While protecting its interests,China should guide global AI security governance back to the path of win-win cooperation.
出处
《国际观察》
CSSCI
北大核心
2024年第4期1-30,共30页
International Review
基金
福建省科技计划项目“面向国家科技安全的多模态开源情报智能应用研究”(项目批准号:2024R1044)的阶段性成果。
关键词
美国人工智能政策
安全议程
大国竞争
美国安全战略
安全治理合作
U.S.Artificial Intelligence Policy
Security Agenda
Major Power Competition
U.S.Security Strategy
Security Governance Cooperation
