**中外专家在京签署《北京AI安全国际共识》：禁止AI自我复制及应用于生化武器领域**
据腾讯科技报道，一场重大的国际会议在北京举行，图灵奖得主约书亚·本吉奥、杰弗里·辛顿、姚期智等数十位全球知名专家共同参与，他们一致签署了由智源研究院发起的《北京AI安全国际共识》。这一共识旨在规范人工智能（AI）技术的发展和应用，保障全球社会的安全与福祉。
《共识》明确了人工智能的“风险红线”和“发展路线”，将AI的自我复制、寻求权力、协助不良行为者以及欺骗行为列为高风险领域。特别是，专家们明确指出，禁止AI技术自我复制，以防止其可能引发的不可控局面。此外，共识还强调，AI技术不得应用于生化武器的研制，以维护全球生物安全和国际和平。
这一举措表明，全球科技界对AI技术的潜在风险有了更深入的认识，并正在积极采取行动，以确保科技进步的同时，避免可能带来的灾难性后果。智源研究院作为发起方，其在AI领域的研究和影响力，无疑为这一共识的权威性和实施力度提供了保障。
此次签署的《北京AI安全国际共识》不仅为全球AI行业的健康发展设定了道德和法律的边界，也为国际社会在未来AI治理中提供了参考框架。各方专家的共同参与和承诺，展现了全球科技界对构建安全、负责任的AI环境的共同责任和决心。
英语如下：
**News Title:** “Global Experts Reach Consensus: Ban on AI Replication and Biochemical Applications for Ensuring Future Safety”
**Keywords:** AI Safety Consensus, Expert Signatories, Ban on Replication
**News Content:**
**Renowned Scholars Worldwide Sign the “Beijing AI International Safety Consensus” in Beijing: Prohibiting AI Self-Replication and Use in Biochemical Weapons**
According to Tencent Technology, a major international conference was held in Beijing, attended by Turing Award winners like Yoshua Bengio, Geoffrey Hinton, and Yao Qizhi, among several other globally renowned experts. They jointly signed the “Beijing AI International Safety Consensus,” initiated by the Baidu Institute of Technology. This consensus aims to regulate the development and application of artificial intelligence (AI) technologies, ensuring global security and well-being.
The “Consensus” outlines the “risk red lines” and “development pathways” for AI, identifying self-replication, seeking power, aiding malefactors, and deception as high-risk areas. Specifically, the experts emphasized the prohibition on AI self-replication to prevent uncontrollable scenarios. Furthermore, it underscores that AI technologies should not be employed in the development of biochemical weapons to safeguard global biosecurity and international peace.
This move demonstrates the科技 community’s deeper understanding of the potential risks associated with AI and their proactive efforts to ensure that technological progress does not lead to catastrophic consequences. As the initiator, the Baidu Institute of Technology’s research and influence in the AI field lend credibility and strength to the enforcement of this consensus.
The “Beijing AI Safety International Consensus,” now signed, sets ethical and legal boundaries for the healthy development of the global AI industry and provides a reference framework for international governance in the future. The collective participation and commitment of experts worldwide reflect the shared responsibility and determination of the global tech community to build a safe and responsible AI environment.
【来源】https://new.qq.com/rain/a/20240318A04IFG00
Views: 1