**中外顶尖专家共签《北京AI安全国际共识》：禁止AI自我复制与涉及生化武器应用**
据腾讯科技报道，近日，一场重量级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、姚期智等数十位中外人工智能领域的权威专家共同签署了一份具有里程碑意义的文件——《北京AI安全国际共识》。这份共识由智源研究院发起，旨在为全球AI发展设定安全准则，防范潜在风险。
共识明确了人工智能的“风险红线”，其中包括禁止AI系统自主复制以及应用于生化武器的研发。专家们一致认为，AI的自我复制能力可能带来无法预见的后果，而将其用于生化武器则可能对人类生存构成严重威胁。此外，共识还强调了禁止AI系统寻求权力、协助不良行为者以及进行欺骗行为，以防止技术被滥用。
这份《北京AI安全国际共识》的出台，标志着国际社会对AI伦理和安全问题的高度重视。它不仅为全球AI研究和应用提供了明确的道德和法律指引，也将对未来的科技创新产生深远影响。中外专家的联合行动，展示了全球范围内对AI安全问题的共识和决心，有望推动建立更为完善和统一的国际AI安全标准。
英语如下：
**Title:** “Global and Chinese Experts Sign the ‘Beijing AI Safety Consensus’: Prohibiting AI Self-Replication and Biochemical Weapon Applications”
**Keywords:** AI Safety Consensus, Expert Signatures, Prohibition on Replication
**News Content:**
According to Tencent Technology, recently, a high-profile international conference was held in Beijing, where Turing Award winners Yoshua Bengio, Geoffrey Hinton, Yao Qizhi, and dozens of other renowned AI experts from China and abroad jointly signed a landmark document – the ‘Beijing AI Safety International Consensus.’ Initiated by the Baidu Institute of Technology, this consensus aims to establish safety guidelines for global AI development and mitigate potential risks.
The consensus outlines the “red lines” for AI, specifically banning AI systems from self-replication and their use in the development of biochemical weapons. The experts agreed that the self-replication capability of AI could lead to unforeseen consequences, while its application in biochemical weapons could pose a severe threat to human existence. Additionally, it emphasizes prohibitions on AI systems seeking power, assisting malicious actors, or engaging in deceptive behavior to prevent misuse of the technology.
The introduction of the ‘Beijing AI Safety International Consensus’ signals the international community’s high level of concern regarding AI ethics and security issues. Not only does it provide clear moral and legal guidance for global AI research and application, but it will also have a profound impact on future technological innovation. The joint action by experts from around the world demonstrates a shared consensus and commitment to AI safety issues, potentially paving the way for more comprehensive and unified international AI safety standards.
【来源】https://new.qq.com/rain/a/20240318A04IFG00
Views: 1