返回新闻
政策AI Understanding 简报

习近平在金砖国家峰会上提出开源人工智能合作

据《日本时报》报道,习近平提议建立金砖国家开源人工智能合作社区,包括大语言模型开发、研讨会和培训。

4 min readRead the original reporting
Source-provided image accompanying Xi pitches open-source AI cooperation at BRICS summit
归因报告来源记录
出版商
japantimes.co.jp
来源链接
japantimes.co.jphttps://www.japantimes.co.jp/news/2026/09/14/world/politics/xi-ai-brics-summit-china-us/
来源类型
新闻媒体的报道——不是第一方文件。

我们无法独立确认的内容: 此声明归因于指定的商店。我们没有根据第一方文件对其进行验证。 (japantimes.co.jp)

背景60 秒内了解这一点

从这里开始

关键术语

计算
训练和运行模型所需的处理资源,通常以 FLOPS 或 GPU 小时来衡量。
数据集
用于训练、验证或测试的结构化或非结构化示例的集合。
测试一下自己AI 模型解释测验

发生了什么

The Japan Times reports that Chinese President Xi Jinping proposed a BRICS “community” for open-source AI cooperation and invited BRICS members to join a Beijing-backed organization involving nearly 30 countries.

The Japan Times reports that Xi spoke in New Delhi on Sunday and said China would lead the creation of a BRICS “community” for open-source AI. According to the report, the proposed cooperation would support the development and application of large language models, as well as AI seminars and training courses for BRICS members.

The report says Xi had invited BRICS nations a day earlier to join a new group established by Beijing and involving nearly 30 countries, which he called the World AI Cooperation Organization. Xi urged BRICS to strengthen AI cooperation and encourage “open source, openness, collaboration and sharing.” The supplied report does not identify the participating countries, provide a formal founding document or independently verify the organizations’ operating arrangements.

来源详情: japantimes.co.jp ↗

为什么这很重要

The proposal places AI development, access and skills training within a diplomatic cooperation framework rather than leaving them solely to commercial companies or individual governments. If implemented, it could give emerging-economy institutions a channel for sharing large-language-model expertise and applications. The report does not establish whether China has released a model, committed funding or offered usable infrastructure. It also does not independently confirm the initiative’s membership, governance, technical standards, safety requirements or practical availability.

The reported initiative could matter because it frames AI capacity-building as an international cooperation project involving model development, deployment and training. That approach could be practically relevant for countries seeking access to AI expertise and infrastructure outside established U.S.- and China-centered commercial channels. However, the source provides no evidence that a model, computing resource, , course or service is currently available to the public or to BRICS members.

The announcement also signals that China is using AI cooperation as part of its broader international engagement. Any effect on the global AI ecosystem will depend on implementation: the licensing terms for any released systems, the openness of the technical stack, the reliability of training support and the safeguards applied to high-impact uses. Those details remain unknown, and the Japan Times account is not independently confirmed in the supplied material.

Interactive Mechanism

互动机制:它实际上是如何运作的

以交互方式探索这一发展背后的基础技术。

System Requirements:
Best ArchitecturePure RAGRecommended pattern
Hallucination RiskVery LowGrounding efficacy
Update Cost$0 (Vector sync)Ongoing maintenance
Core takeaway: Fine-tuning teaches models how to speak (form, style, syntax); RAG teaches models what to say (verifiable facts). Never use fine-tuning alone for factual memory.
交互式概念检查+10 Points
AI Models Explained Quiz

Which component of an AI application is the machine-learning model itself?

接下来看什么

The key tests will be whether the proposed community receives a formal charter, publishes membership and governance details, releases models or tools under verifiable open-source licenses, and provides the promised seminars or training. No access terms, pricing, launch schedule, funding commitments or model specifications are documented in the supplied report. It is also unclear how the initiative would interact with export controls, existing international AI organizations and national safety rules.

A formal membership list and charter for the proposed BRICS community and World AI Cooperation Organization would clarify whether these are operating institutions or initiatives still being organized. The report does not state when either group will begin work.

Any subsequent model or tool release should be assessed for its license, technical documentation, requirements, language coverage, safety testing and actual availability. The supplied source gives no price, access mechanism, performance result or release date.

Governments’ positions on data sharing, export controls, training standards and accountability will show whether the proposal becomes a practical AI cooperation mechanism. The report does not describe reactions from other BRICS members or independent evaluations of China’s plans.

相关指南和测验

人工智能模型解释AI 的未来AI 伦理测试你所知道的——尝试免费的人工智能测验在我们的词汇表中查找人工智能术语关注AI监管追踪器
觉得这有用吗?