德国研究人员推出了一种由极薄半导体层制成的超快光开关，其运行速度比当今的电子晶体管快约 10,000 倍。 由奥尔登堡大学物理学家团队设计的新型光学器件，可用作光开关或光晶体管。研究人员表示，超快光开关为光数据处理提供 了广阔的应用前景。 这种被描述为纳米结构"活性超材料"的原型由银和原子级薄的半导体层制成。它能够在飞秒（即千万亿分之一秒）的时间尺度 上控制光。 "如果我们想在纳米尺度上制造超快光开关，我们的发现就特别有意义，"奥尔登堡大学实验物理学教授、该研究的主要作者克 里斯托夫·利瑙博士表示。 纳米尺度光学 利瑙和他的研究团队着手寻找一种材料，利用聚焦激光束，可以在短短几个飞秒内操控或切换其反射特性。作为参考，一飞秒 等于十亿分之一秒的百万分之一。 在这个项目中，研究团队使用了一种超薄的银纳米狭缝阵列。然后，他们在其表面铣削出一个平行沟槽网格，每个沟槽的宽度 和深度约为 45 纳米（十亿分之一毫米）。 参与这项研究的英国剑桥大学的科学家们随后将一层仅有三个原子厚的半导体晶体二硫化钨单层涂覆到该结构的表面上。 莫里茨·吉廷格博士（左）和丹尼尔·蒂默博士（右）站在这项研究所依据的实验装置前。图片来源：奥尔登 ...

现在AI圈卷算力卷得比奶茶店卷新品还疯狂，但咱们总被"卡脖子"闹心，直到这款中国"算力黑马"横空 出世！ 中国LightGen全光芯片直接碾压A100，100倍算力buff叠满，换道反超美国，这波操作简直帅到爆炸！ 这枚仅指甲盖大小的芯片，却干了大事。 不仅同时攻克三大世界级技术难题，更给国家的算力焦虑，交出了最硬核的中国方案！ 先给大家划重点：现在 AI 算力需求涨得比金价还猛，传统电子芯片早就逼近摩尔定律的"物理天花 板"了，光刻机卡脖子、能耗高到像"电老虎"，这些都是咱们发展的"拦路虎"。 而美国正是抓住这一点，在电子芯片产业链上层层设限。 但他们万万没想到，我们早已开辟全新赛道——光计算！ 更关键的是，团队用薄膜铌酸锂材料，解决了串扰和热漂移的老毛病，让这么多神经元能稳定协同工 作，这就为处理复杂的AI生成任务，打下了最坚实的硬件基础。 用光子代替电子做计算，天生就自带"高速、并行、低功耗"的主角光环，这就是中美芯片竞争的"换道 逻辑"。 LightGen一出场，直接让我们在这条新赛道上，从"跟跑小弟"逆袭成"领跑大哥"，这波反杀太解气了！ 可能有人会问，光计算不是早就有人研究吗？ 没错，但全球范围内 ...

来源：环球网 【环球网科技综合报道】1月11日消息，据《先进光子学》（Advanced Photonics）报道，在人工智能神 经网络高速发展的背景下，大规模的矩阵运算与频繁的数据迭代给传统电子处理器带来了巨大压力。光 电混合计算通过光学处理与电学处理的协同集成，展现出显著的计算性能，然而实际应用受限于训练与 推理环节分离、离线权重更新等问题，造成信息熵劣化、计算精度下降，导致推理准确度低。 中国科学院半导体研究所提出了一种基于相位像素阵列的可编程光学处理单元（OPU），并结合李雅普 诺夫稳定性理论实现了对OPU的灵活编程。在此基础上，团队构建了一种端到端闭环光电混合计算架构 （ECA），通过硬件—算法协同设计，实现了训练与推理的全流程闭环优化，有效补偿了信息熵损失， 打破了光计算中计算精度与准确度之间的强耦合关系。 该架构通过噪声自学习机制，实现了光学与电学参数联合优化与自适应计算精度补偿。实验结果表明， 采用4-bit的OPU时，ECA在MNIST手写数字识别任务（计算机视觉领域的经典任务）上的推理准确率达 到90.8%，接近8-bit传统计算架构（TCA）的理论极限（90.9%），这表明光计算系统在低硬 ...

Core Viewpoint - The company Guangbenwei Technology is developing glass-based optical computing chips to replace silicon substrates, aiming to achieve a new era of "1000 tops computing power and 1000 tops/W energy efficiency" for AI applications [1][2]. Group 1: Technology and Innovation - The glass optical computing chips are designed for AI inference scenarios, with the market expected to reach $255 billion by 2030, as inference will account for 75% of total AI computing [1]. - The company has completed its first commercial-grade optical computing chip, with a matrix size of 128mm x 128mm and a peak computing power exceeding 1000 tops [1]. - Glass is chosen over silicon due to its superior properties such as flatness, thermal stability, and compatibility with optical waveguide processes, allowing for larger chip designs and improved performance [2]. Group 2: Performance and Efficiency - The glass optical computing chip can achieve a computing power of 2600 tops when sized at 200mm x 200mm, significantly enhancing computational capabilities [2]. - The company has developed a zero static power consumption feature for its optical computing chips, predicting an energy efficiency ratio exceeding 1000 tops/W, which is over 200 times that of TPU [2]. Group 3: Future Developments - Guangbenwei Technology aims to create a next-generation all-optical computing system that allows light signals to perform repeated calculations and dynamic storage, moving beyond the limitations of single-core optical computing [3]. - The ultimate goal is to package the glass optical computing chip into a high-performance all-optical computing system, defining the next generation of AI computing standards and providing comprehensive solutions for various user scenarios [3].

Core Insights - The article discusses the recent surge in AI-related companies listing in Shanghai, highlighting the city's appeal as a hub for talent and innovation in the AI industry [1][3]. Talent Ecosystem - Shanghai is recognized as one of the most attractive cities for international talent in China, drawing many young entrepreneurs in the tech sector [1]. - Shanghai Yiyu Technology, established in 2022, has a team of 385 members with an average age of 29, and over 70% of its revenue comes from international markets [1]. - Shanghai Tensu Zhixin Semiconductor Co., which went public on January 8, 2023, also cites the concentration of top chip design talent in Shanghai as a key reason for its location choice [1]. Scene Implementation - Shanghai hosts over 1,200 integrated circuit companies, accounting for approximately 40% of China's related industry talent, with more than 40 companies listed across various capital markets [3]. - Mu Xi Integrated Circuit (Shanghai) Co., founded in 2020, is set to list on the Shanghai Stock Exchange's Sci-Tech Innovation Board in December 2025 [3]. - The open and fair environment in Shanghai allows technology-focused companies to concentrate on innovation and quickly iterate their products [3]. Cutting-edge Innovation - Data from the Shanghai Municipal Economic and Information Commission indicates that the AI industry in Shanghai is thriving, with 394 large-scale enterprises generating revenue of 435.492 billion yuan (approximately 61.5 billion USD) in the first three quarters of 2025, marking a year-on-year growth of 39.6% [4]. - The demand for computing power is rapidly increasing, and Shanghai is solidifying its technological foundation for industrial innovation, with AI chip companies like Xizhi Technology emerging [4]. - Xizhi Technology's CEO believes that optical computing could provide a differentiated path for domestic computing power, akin to the transition from fuel vehicles to electric vehicles [5][6].

Group 1: AI Developments - DeepSeek's open-source model DeepSeek-R1 utilizes pure reinforcement learning, significantly reducing the dependency on labeled data while achieving top performance under limited computational power [2] - A new brain-computer interface developed by the University of Texas at Austin can decode thoughts into continuous text in about one hour, showcasing a significant leap in efficiency and applicability [3] - AI has been used to design complex serine hydrolases from scratch, marking a key advancement in computational biology and understanding life’s catalytic mechanisms [4][5] Group 2: Quantum Computing - Google achieved a significant breakthrough in quantum computing by demonstrating verifiable quantum advantage with its "Willow" quantum processor, completing a task 13,000 times faster than classical supercomputers [10][11] Group 3: Astronomy and Space Exploration - The Vera C. Rubin Observatory released its first test images, showcasing its capability to capture millions of distant stars and galaxies, marking a transformative step in astronomical observation [7] Group 4: Robotics in Medicine - A new intelligent robot successfully performed a complete gallbladder removal surgery autonomously, demonstrating a significant advancement in surgical robotics by integrating high precision with adaptive understanding [8][9] Group 5: Climate Change and Environmental Science - The Global Critical Points Report indicates that the world is approaching several catastrophic climate thresholds, with the first significant sign being the mass death of warm-water coral reefs due to temperature increases [13][14] Group 6: Neuroscience - A comprehensive cross-species mammalian brain cell development map was published, revealing critical insights into the development and function of brain cells, which could lead to better understanding and intervention for neurodevelopmental disorders [15][16]

Group 1: Market Performance - The ChiNext AI ETF (159242) rose by 3.12% with a turnover rate of 11.95%, and a transaction volume of 36.43 million yuan, indicating active market trading [1] - The underlying index, ChiNext AI Index, increased by 3.24%, with notable gains from constituent stocks such as Zhishang Technology (+13.73%), Changxin Bochuang (+13.10%), and Xinyi Sheng (+6.19%) [1] Group 2: Industry Focus - The ChiNext AI Index emphasizes AI engineering and industrialization, focusing on foundational technology sectors like optical modules, computing chips, edge computing, and operating systems, distinguishing itself from other tech indices that prioritize algorithm models [1] - The top ten holdings of the index include companies like Xinyi Sheng, Zhongji Xuchuang, and Tianfu Communication, which are centered on domestic AI innovation and application [1] Group 3: Technological Advancements - Research from Shanghai Jiao Tong University introduced the LightGen all-optical computing chip, which supports large-scale semantic visual generation models, marking a breakthrough in next-generation computing chips [2] - Light computing is highlighted for its scalability, low power consumption, ultra-high speed, and high parallelism, making it crucial for AI, scientific computing, and large-scale data processing [2] Group 4: Future Projections - According to LightCounting's report, the global market for optical transceivers and related products is expected to exceed $23 billion by 2025, reflecting a significant 50% year-over-year growth [2] - The sales of Ethernet optical modules for data center interconnects are projected to reach $17 billion, a 60% increase compared to the previous year [2] - The introduction of the new GPU architecture "Huagang" by Moore Threads aims to enhance computing density by 50% and energy efficiency by ten times, supporting large-scale intelligent computing clusters [3]

Core Viewpoint - The semiconductor industry chain in Hong Kong is experiencing significant growth, with notable increases in stock prices for key companies and the launch of a new ETF focused on this sector [1][3]. Group 1: Market Performance - On December 22, major semiconductor stocks in Hong Kong saw substantial gains, with SenseTime-W rising over 7%, SMIC increasing nearly 6%, and Hong Teng Precision up nearly 5% [1]. - The first ETF in the market focusing on the Hong Kong semiconductor industry, the Hong Kong Information Technology ETF (159131), recorded a price increase of 1.69% and a trading volume exceeding 30 million yuan [1]. Group 2: Technological Breakthroughs - Researchers from Shanghai Jiao Tong University achieved a breakthrough in optical computing chips, integrating over 1 million optical neurons on a single chip, which enhances AI capabilities by reducing latency and energy consumption [3]. - This advancement addresses critical issues in the AI industry, such as high costs and slow training times, and offers a transformative path for real-time AI and green AI applications [3]. Group 3: ETF Composition and Strategy - The Hong Kong Information Technology ETF is structured with a focus of 70% on hardware and 30% on software, heavily investing in semiconductor, electronics, and computer software sectors [3]. - The ETF includes 42 Hong Kong hard-tech companies, with significant weights assigned to SMIC (20.48%), Xiaomi Group-W (9.53%), and Hua Hong Semiconductor (5.80%), while excluding major internet firms like Alibaba and Tencent [3].

Group 1 - The A-share market saw all three major indices open higher on December 22, with the Sci-Tech 50 ETF (588000) experiencing a maximum increase of 1.47% during early trading [1] - The top-performing stocks included Tuojing Technology, which rose by 8.50% to reach a historical high, benefiting directly from increased demand for optical computing chips and advanced process equipment [1] - The Sci-Tech 50 ETF has attracted significant capital inflow, with a net inflow of 1.787 billion over the past five days and 2.644 billion over the past ten days, indicating strong investor interest [1] Group 2 - According to招商证券, the growth in AI demand is expected to drive global storage and advanced process capacity expansion, with domestic storage and advanced process expansion likely to accelerate by 2026-2027 [2] - The report highlights that domestic equipment manufacturers are seeing a positive trend in orders, and companies with strong market positions in storage equipment are expected to benefit from increasing domestic production rates [2] - The Sci-Tech 50 ETF tracks the Sci-Tech 50 Index, with 69.39% of its holdings in the electronics sector and 4.88% in the computer sector, aligning well with the development of cutting-edge industries such as AI and robotics [2]

Group 1: Industry Developments - A Starlink satellite from SpaceX experienced an anomaly, losing contact with the ground and producing debris, which poses a potential threat to other satellites in orbit [2] - The first L3-level autonomous driving license in China was issued to Changan Automobile, marking a significant step towards commercial application of L3 autonomous driving technology [3] - Beijing's humanoid robotics center has open-sourced the XR-1 model, which is the first to pass national standards for embodied intelligence, advancing the domestic embodied intelligence industry [4] - Turing Quantum and Moore Threads have signed a strategic cooperation agreement to develop a hybrid computing architecture that integrates GPU and quantum processors [5] - Chinese robot "Deep Sea Giant Shark 3" won a combat robot competition in Abu Dhabi, showcasing advanced design and capabilities [6] - Researchers at Shanghai Jiao Tong University achieved a breakthrough in optical computing chips, supporting large-scale semantic media generation models [7] Group 2: Corporate News - Xiaomi's president confirmed that the price of the upcoming Xiaomi 17 Ultra smartphone will increase due to rising memory costs, which are expected to continue increasing in the coming years [2] - New Unisplendour Group signed a strategic cooperation agreement with Shanghai Yidian to collaborate on artificial intelligence and intelligent computing [10] - Wallen Technology plans to acquire at least 60% of Liao Jing Electronics, which operates in integrated circuits and electronic components [14] - Tianpu Co. announced the expiration of a tender offer from Zhonghao Xinying for 33.52 million shares, representing 25% of the company's total equity [15] - Lianchuang Electronics is planning a change in control due to a potential transfer of shares by its major shareholder [16] - *ST Dongyi's restructuring plan was approved by the court, allowing the company to focus on its main business while leveraging resources in the computing industry [17]