北京，中国 – 据腾讯科技报道，在近日举行的一次国际会议上，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，共同签署了《北京AI安全国际共识》。该共识由智源研究院发起，旨在为人工智能的发展设置“风险红线”和明确“发展路线”，以确保AI技术的健康发展，防止其被滥用。
《北京AI安全国际共识》中明确了人工智能的四大“风险红线”，包括禁止AI自行复制、禁止将AI应用于生化武器等领域。此外，共识还提出了对AI权力滥用的防范措施，以及禁止AI协助不良行为者和欺骗行为。
专家们强调，人工智能作为一种强大的技术力量，必须在其发展的初期就建立起风险控制机制，以防止技术失控带来的灾难性后果。本次签署的共识，被视为国际社会在AI治理领域的一次重要协作，为全球AI安全规范的制定提供了中国方案。
此次共识的签署，体现了国际社会对于人工智能安全的共同关注，以及对于负责任AI发展的共识。据了解，该共识的发布将引导全球AI行业的健康发展，同时也为相关政策的制定提供了科学依据和技术指南。
作为一项国际性的共识，它不仅标志着全球AI治理体系的进一步完善，也展现了中国在AI领域内负责任的态度和领导力。未来，随着更多国家和组织的加入，相信《北京AI安全国际共识》将成为全球AI治理的重要参考，共同守护人工智能技术的可持续发展。
英语如下：
Certainly, here is the translation in English following the markdown format:
“`markdown
# Beijing AI Safety Consensus Signed by Experts at Home and Abroad: AI Prohibited from Self-Replication
Keywords: AI risk redline, Beijing Consensus, Expert Signing.
### Beijing AI Safety International Consensus Released: Establishing “Traffic Lights” for Artificial Intelligence Development
Beijing, China – According to Tencent Technology, at a recent international conference, dozens of well-known experts from China and abroad, including Turing Award winners Joshua Benigo, Geoffrey Hinton, and Yao Qizhi, jointly signed the “Beijing AI Safety International Consensus”. The consensus was initiated by the Beijing Academy of Artificial Intelligence (BAAI) with the aim of setting “risk redlines” and clarifying “development routes” for the healthy development of artificial intelligence and preventing its abuse.
The “Beijing AI Safety Consensus” identifies four major “risk redlines” for artificial intelligence, including the prohibition of AI self-replication and the prohibition of using AI in the field of biological and chemical weapons. In addition, the consensus proposes preventive measures against AI power abuse and bans AI from assisting bad actors and deceptive behaviors.
Experts emphasize that artificial intelligence, as a powerful technological force, must establish risk control mechanisms in its early stages of development to prevent catastrophic consequences from out-of-control technology. The signing of this consensus is seen as an important international collaboration in the field of AI governance and provides a Chinese approach to the formulation of global AI safety standards.
The signing of this consensus demonstrates the international community’s shared concern for AI safety and a consensus on responsible AI development. It is understood that the release of this consensus will guide the healthy development of the global AI industry and provide a scientific basis and technical guidelines for policy formulation.
As an international consensus, it not only marks the further improvement of the global AI governance system but also shows China’s responsible attitude and leadership in the AI field. In the future, with the participation of more countries and organizations, the “Beijing AI Safety Consensus” is expected to become an important reference for global AI governance and collectively safeguard the sustainable development of artificial intelligence technology.
“`
This translation maintains the structure and emphasis of the original Chinese text while adapting it for an English-speaking audience.
【来源】https://new.qq.com/rain/a/20240318A04IFG00
Views: 5