**中外顶尖专家共签《北京AI安全国际共识》：划定AI技术安全边界**
日前，一场国际性的AI安全研讨会在京举行，吸引了包括图灵奖得主约书亚·本吉奥、杰弗里·辛顿、姚期智等在内的数十位中外知名专家的参与。在此次会议上，专家们一致签署了一份由智源研究院发起的《北京AI安全国际共识》，旨在为全球人工智能的发展设定明确的安全准则。
该共识明确提出，人工智能技术的发展必须遵守一系列“风险红线”，其中最为引人关注的是禁止AI系统自行复制以及将其应用于生化武器领域。此外，共识还强调，AI不得寻求自主权力、协助不良行为者或进行欺骗行为，以防止技术滥用对社会造成潜在危害。
《北京AI安全国际共识》的出台，反映了全球科技界对于AI伦理和安全问题的深度关切。这一文件不仅为AI的研发和应用设定了明确的道德与法律边界，同时也为各国政府和企业提供了指导性的原则，以确保人工智能在推动社会进步的同时，不会对人类安全和福祉构成威胁。
智源研究院作为此次共识的发起方，表示将持续推动AI安全的研究与国际合作，以期在全球范围内建立起更加完善的人工智能安全管理体系。这一举措无疑为全球AI领域的健康发展树立了新的标杆，也为未来科技与社会的和谐共存提供了有力的保障。
英语如下：
**News Title:** “Global Experts Reach Consensus: Ban on AI Replication and Biochemical Applications for Future Safety”
**Keywords:** AI Safety Consensus, Expert Signatories, Ban on Replication
**News Content:**
**Top International Experts Sign the “Beijing AI Safety International Consensus”: Setting Boundaries for AI Technology Safety**
Recently, an international AI safety symposium was held in Beijing, attracting dozens of renowned experts, including Turing Award winners Yoshua Bengio, Geoffrey Hinton, and Yao Qizhi, among others. During the conference, the experts jointly signed the “Beijing AI Safety International Consensus,” initiated by the Institute of Intelligence Source, to establish clear safety guidelines for global AI development.
The consensus explicitly outlines a series of “risk red lines” that AI technology must adhere to. Most notably, it prohibits AI systems from self-replicating and their application in biochemical weaponry. It also emphasizes that AI should not seek autonomy, assist malicious actors, or engage in deceptive behavior to prevent potential harm from technological misuse.
The emergence of the “Beijing AI Safety International Consensus” reflects the global tech community’s deep concern over AI ethics and safety issues. This document not only sets moral and legal boundaries for AI research and application but also provides guiding principles for governments and companies worldwide to ensure that AI advancements promote societal progress without threatening human safety and well-being.
As the initiator of the consensus, the Institute of Intelligence Source vowed to continue promoting AI safety research and international cooperation, aiming to establish a more robust global AI safety management system. This initiative sets a new benchmark for the healthy development of the global AI field and provides strong assurance for the harmonious coexistence of future technology and society.
【来源】https://new.qq.com/rain/a/20240318A04IFG00
Views: 1