Google Cloud Speech-to-Text API (Chirp) + Gemini 2.5 Flash Non-Thinking. Treat it as a searchable index of what was broadcast, not a quotation record.
00:02:02Hello, friends and viewers, welcome to Focus Interview. At the 2026 World Artificial Intelligence Conference and the High-level Conference on Artificial Intelligence Global Governance,
00:02:06会 及 人 工 智 能 全 球 治 理 高 级 别 会 议 开 目 市 上
00:02:09President Xi Jinping proposed to jointly build a fair and reasonable
00:02:13global artificial intelligence governance system. The first point is
00:02:17to adhere to open co-operation to drive innovation and development,
00:02:20and to encourage open source collaboration and sharing.
00:02:23Around this conference, domestic open-source large models were intensively launched.
00:02:27KMK3 with
00:02:3128,000 units
00:02:34and Alibaba's Deepsick, etc.
00:02:38Open source is becoming the most prominent feature of China's artificial intelligence.
00:02:44July 27, 2026, the new era of artificial intelligence.
00:02:45月27號人工能越新期
00:02:50The launch of KimyK3 is not the first open-source of a large Chinese model, but it can produce a chess piece of 28,000 digits.
00:03:30What's even more surprising is that this open-source not only includes
00:03:34the full weight of the model, but also technical reports and three underlying
00:03:38model training methods. What is open
00:03:41source? We can compare a large model to a restaurant. Using
00:03:45a closed-source model is like going to a restaurant to eat. The food is delicious, but you can't
00:03:49take it home. If you want to eat, you can only come to the restaurant. But using an open-source model is
00:03:52like sharing the secret of your restaurant. You can come to the restaurant to eat,
00:03:56or you can take the ingredients home and make it yourself. You can also.
00:04:33Kimi K3 was launched as an open-source project and immediately ranked first globally in the
00:04:36Fronten and Code
00:04:37Arena with 1679 points.
00:04:44At the World Artificial Intelligence Conference on July 17th, although there was only a simple video introduction,
00:04:49Kimi's booth still attracted many domestic and foreign visitors who came for technical exchanges or experiences.
00:04:57Kimi launched K3, such a big
00:05:33chip design animation made by K3.
00:05:35You might think K3 only made such an animation, but actually it's not. This is just K3 for chip design.
00:06:32Such a chip design task usually requires multiple experienced chip designers to collaborate, but K3 only received
00:06:35the simplest prompt, and independently completed the entire process.
00:06:39It can independently acquire these.
00:06:54The release of KMK3 is just one of the many new updates recently.
00:07:00Since July this year, China's open source has been upgraded intensively.
00:07:06On July 31st, Deepsiek released V4
00:07:09flash official version, its capabilities are greatly
00:07:15Deep seek V4 Pro official version.
00:07:19On August 3rd, 208 launched a large model
00:07:233.8
00:07:23max 24,000 units, then open source.
00:07:29it is the second largest open-weighted model in China
00:07:32so far, just behind K3 and according to
00:07:35Alibaba, it trails just behind anthropic on certain
00:07:39benchmarks on the arena leerboard and it even
00:07:50The spring report of 2026 shows that the download volume of open-source models developed in China
00:07:54accounts for 41% of the global total, surpassing the United States
00:07:58to rank first in the world. On the global mainstream large model invocation
00:08:01list, the top five ranked models all come from Chinese
00:08:05teams. In the past 12 months, domestic models have maintained
00:08:09the upper limit of the global open-source model scale for nine months.
00:09:16China is using
00:09:18its own way to achieve modernization.
00:09:21.
00:09:26Keep up the good work.
00:09:27.
00:09:31Openness and inclusiveness are the core attitudes of China's open-source large models.
00:09:34It is precisely this attitude that allows open-source large models to quickly enter
00:09:38enterprises, boost production, and transform into tangible and visible
00:09:42real productivity. Product testing is currently being conducted on this screen.
00:09:46In just a few seconds, defects on the
00:09:50product were identified by AI and a repair plan was provided,
00:09:54and the machine tool was directly controlled to complete the repair.
00:09:57In the past, this identification process was performed manually,
00:10:01which was less efficient. This AI system's technical foundation
00:10:04is called the Lan Jing Display Large Model,
00:10:07developed and designed by Jingdongfang, but its base model
00:10:11is from Cai Na.
00:10:25Mismatch, open-source models have this advantage.
00:10:28Our large model is deployed in our enterprise's domain. In this way, our
00:10:32information security and data security can be guaranteed to the greatest extent.
00:10:36Lower cost, deep customization, private deployment.
00:10:39These three advantages allow manufacturing enterprises like Jingdongfang
00:10:43to transform general open-source large models into industry-specific AI brains
00:10:47through secondary training for their own use. Currently, the Lan
00:10:51Jing Display Large Model has been implemented in Jingdongfang's nationwide
00:13:25How can such a good technology, or a breakthrough technology,
00:13:28be effectively used? This is a question that China has always been concerned about.
00:13:32Therefore, the views put forward by President Xi Jinping in Shanghai this time
00:13:36actually precisely address this concern. On July 17, 2026,
00:13:40President Xi Jinping
00:13:44delivered a keynote speech at the 2026 World Artificial Intelligence Conference and the High-level Conference on Artificial Intelligence Global Governance,
00:13:47全 球 治 理 高 级 别 会 议 开 幕 市 上 发 表 主 指 讲 话
00:13:51proposing to jointly build a fair and reasonable global.
00:13:57The first point is to adhere to open co-operation,
00:14:01driving innovation and development. China advocates open cooperation
00:14:05and a governance concept for mutual development, which highly aligns with the development laws of the open-source ecosystem.
00:14:09Openness does not mean laissez-faire.
00:14:13The core is to establish a governance mechanism for the development of adaptive technologies,
00:14:17achieving a dynamic balance between innovative vitality and risk prevention and control.
00:14:20Beyond concepts, there are actions. As early as
00:14:242023, China took the lead in releasing the Global Artificial Intelligence Governance
00:14:27Initiative, clearly advocating a people-centered and AI for good governance philosophy.
00:14:31On July 17, 2026, China's Ministry of Industry and Information Technology,
00:14:35together with relevant parties, jointly guided the release of the International Artificial Intelligence
00:14:39Ethics Governance Action Plan, focusing on ethical governance throughout the entire lifecycle of artificial intelligence,
00:14:42risk classification and grading prevention and control, flexible
00:14:46governance mechanism construction and other directions, urging all countries to strengthen policy
00:14:50coordination and practical cooperation. On July 16,
00:14:54the establishment of the World Artificial Intelligence Cooperation Organization Agreement signing ceremony
00:14:58was held in Shanghai. In the next stage, the core question
00:15:01that truly needs to be answered is how to effectively identify,
00:15:05prevent, and mitigate new risks brought by artificial intelligence
00:15:08while promoting technological innovation, and to explore a dynamic balance between development and security.
00:15:12Roughly speaking, the answer can be approached from three directions:
00:15:16first, promoting the coordinated development of innovation and security governance.
00:15:19Second, how to implement classified and graded
00:15:23approaches based on risk levels.
00:15:38From domestic open-source large models deeply empowering various industries, to China
00:15:42proposing to build a fair and reasonable global artificial intelligence governance system
00:15:45with four points of view, China is demonstrating through practical actions
00:15:48that the future of artificial intelligence cannot be separated from the power of global
00:15:52open collaboration. For the world,
00:15:55only by upholding open co-operation and promoting development and security,
00:15:59can artificial intelligence better serve all
00:16:02humanity. Thank you for watching today's Focus Interview. Goodbye.
00:16:05.
00:16:38News background analysis.
00:16:39.
00:16:42In-depth analysis of social hot spots.
00:16:44.
00:16:47Commentary on popular topics.
00:16:49.
00:16:52Please watch Focus Interview.
00:16:54.
00:21:50Village development, everywhere vibrant.
00:21:53.
00:21:56Beautiful villages like paintings, struggle. .