单词学习：![[81e13391-e839-4dda-8177-122c1d3c5af9.mp3]]

新闻播报：![[14346512-b5a9-4964-8004-218caeed4da1.mp3]]

千问模型登顶全球开源榜，OpenAI被指发展放缓

简报：

• 中国的开源大模型“千问”（Qwen）在全球范围内的下载量已登顶，成为最受欢迎的开源模型之一。
• 与此同时，业界有观点认为，以OpenAI为代表的闭源模型发展模式正面临挑战，其发展速度被指出现放缓或“失速”现象。

相关链接：

• 中国开源大模型（千问）下载量全球登顶，OpenAI被指“封闭失速”

Qwen Model Climbs to Top of Global Open-Source Rankings 🌐, OpenAI Faces Claims of Slowing Progress

Brief:

• China's open-source large model, "Qwen," has climbed to the top of global download charts, establishing itself as one of the most popular open-source models worldwide. 🚀
• Concurrently, industry observers suggest that the closed-source model development approach, exemplified by OpenAI, is facing challenges, with its development speed reportedly slowing down or even "stalling." 📉

Related Link:

• Chinese Open-Source Large Model (Qwen) Tops Global Downloads, OpenAI Accused of "Closed-Source Stagnation"

OpenAI发力教育领域，为美高校提供70万份ChatGPT许可

简报：

• 根据资料，OpenAI向美国高校提供了70万份ChatGPT许可，意在加速其在教育领域的战略布局。

相关链接：

• OpenAI向美国高校提供70万份ChatGPT许可，加速教育领域布局

OpenAI Pushes into Education Sector, Provides 700,000 ChatGPT Licenses to US Universities 🚀

Brief:

• According to sources, OpenAI has supplied 700,000 ChatGPT licenses to US higher education institutions, aiming to accelerate its strategic expansion in the education field. 📚🎓

Related Link:

• OpenAI向美国高校提供70万份ChatGPT许可，加速教育领域布局

ChatGPT安卓版更新，Plus用户可调节AI思考时长

简报：

• OpenAI更新了安卓版ChatGPT应用，新增“思考时长”（Thinking time）功能，此前该功能仅限于桌面端。
• 用户可在两种模式间切换：“标准思考”用于快速响应，“扩展思考”则投入更长时间进行深度逻辑推演，以处理复杂的数学或编程问题。
• 此项新功能目前是ChatGPT Plus订阅会员的专属特权。

相关链接：

• OpenAI 更新手机版 ChatGPT，现支持手动调节 AI“思考深度”

ChatGPT Android App Update: Plus Users Can Adjust AI Thinking Time 🤖

Brief:

• OpenAI has updated its ChatGPT Android app, introducing a "Thinking time" feature previously exclusive to the desktop version. 📱
• Users can switch between two modes: "Standard Thinking" for quick responses, and "Extended Thinking" which dedicates more time to deep logical deduction, designed for complex mathematical or programming problems. 🧠
• This new feature is currently an exclusive privilege for ChatGPT Plus subscribers. ✨

Related Link:

• OpenAI Updates Mobile ChatGPT, Now Supports Manual Adjustment of AI "Thinking Depth"

哈工大联合博实股份发布双形态人形机器人天机

简报：

• 哈尔滨工业大学与博实股份联合发布了一款名为“天机”的人形机器人。
• 该机器人具备独特的“双形态”结构，能够在人形的两足行走和四轮驱动的轮式移动之间进行切换，以适应不同环境。
• 机器人由团队自研的“天元”大模型驱动，使其能够理解复杂的自然语言指令并自主执行任务。

相关链接：

• 哈工大与博实股份发布由自研大模型驱动的双形态人形机器人

HIT and BoShi Co., Ltd. Jointly Release Dual-Mode Humanoid Robot "Tianji"

Briefing:

• Harbin Institute of Technology (HIT) and BoShi Co., Ltd. have jointly unveiled a humanoid robot named "Tianji."
• This robot features a unique "dual-mode" structure, capable of switching between bipedal humanoid walking and four-wheel drive wheeled movement to adapt to various environments. 🚶‍♂️⚙️
• Powered by the team's self-developed "Tianyuan" large model, the robot can understand complex natural language instructions and autonomously execute tasks. 🤖🧠

Related Link:

• HIT and BoShi Co., Ltd. Release Dual-Mode Humanoid Robot Driven by Self-Developed Large Model

微软Copilot免费集成GPT-5.2，新增“专家级”办公模式

简报：

• 微软已向网页版、Windows 11版和移动端Copilot用户全量免费推送GPT-5.2模型，并将其命名为“智能 +”（Smart Plus）模式。
• 该版本被官方定义为能胜任特定办公任务的“专家级”工具，显著增强了处理电子表格、演示文稿、代码编写和图像处理等复杂任务的能力。
• 在一项针对知识型工作的评估中，GPT-5.2 Thinking在70.9%的案例中表现击败或持平人类行业专家，远超前代模型的38.8%。
• 在软件工程测试SWE-Bench Pro、美国数学邀请赛AIME 2025等多项基准测试中，GPT-5.2均展现出优于前代模型的性能。

相关链接：

• 微软 Copilot AI 应用免费上线 GPT 5.2，“专家级”处理复杂办公任务

Microsoft Copilot Integrates GPT-5.2 for Free, Adds "Expert-Level" Office Mode

Briefing:

• Microsoft has rolled out the GPT-5.2 model completely free to all Copilot users across web, Windows 11, and mobile platforms, naming it the "Smart Plus" mode. ✨
• This version is officially defined as an "expert-level" tool capable of handling specific office tasks, significantly enhancing its ability to tackle complex assignments like spreadsheets, presentations, code writing, and image processing. 🚀
• In an evaluation focusing on knowledge-based work, GPT-5.2 Thinking performed at or above human industry experts in 70.9% of cases, vastly outperforming the previous model's 38.8%.
• GPT-5.2 has demonstrated superior performance compared to its predecessors in multiple benchmark tests, including the SWE-Bench Pro software engineering test and the American Invitational Mathematics Examination (AIME) 2025. 🧠

Related Links:

• Microsoft Copilot AI App Rolls Out GPT 5.2 for Free, Offers "Expert-Level" Handling of Complex Office Tasks

AI重塑就业市场，斯坦福CS学位光环不再

简报：

• 据报道，因生成式AI技术崛起，曾被视为“黄金入场券”的斯坦福大学计算机科学学位价值大跌，毕业生在知名科技公司寻找入门级工作变得困难。
• 企业招聘经理称，由于AI在基础编程任务上的速度和准确率已超越初级程序员，过去需十名工程师的工作现在仅需“两名资深工程师加AI”即可完成，导致校园招聘规模大幅缩减。
• 斯坦福大学的一项研究显示，自2022年底以来，22至25岁初级软件开发人员的就业率已下降近20%。
• 这一趋势已是全球现象，英国等国的IT就业市场也出现类似趋势，同时客户服务、编辑和会计等职业约40%的任务也面临被AI替代的风险。

相关链接：

• 报告称斯坦福计算机学位“黄金入场券”褪色，初级程序员就业率暴跌 20%

AI Reshaping Job Market, Stanford CS Degree Losing Its Luster 📉

Brief:

• According to reports, due to the rise of generative AI technology, a Stanford University Computer Science degree, once considered a "golden ticket," has seen its value plummet. Graduates are now finding it difficult to secure entry-level positions at prominent tech companies.
• Corporate hiring managers state that AI's speed and accuracy in basic programming tasks now surpass that of junior programmers. What previously required ten engineers can now be accomplished by "two senior engineers plus AI," leading to a significant reduction in campus recruitment. 🤖
• A study by Stanford University indicates that since late 2022, the employment rate for junior software developers aged 22 to 25 has fallen by nearly 20%.
• This trend is a global phenomenon, with similar shifts observed in IT job markets in countries like the UK. Concurrently, approximately 40% of tasks in professions such as customer service, editing, and accounting also face the risk of AI automation. 💼

Related Links:

• Report says Stanford computer science degree's "golden ticket" fades, employment rate for junior programmers plummets 20%