➀ OpenAI, the creator of ChatGPT, is about to finalize its first in-house AI chip design;

➁ The company plans to send the chips to TSMC for fabrication with an aim for mass production in 2026;

➂ This move is a significant step towards reducing reliance on NVIDIA.

➀ AZoNetwork joins the global effort to close the gender gap in STEM;

➁ Women in STEM are gaining recognition for their groundbreaking contributions;

➂ Female scientists discuss the future of Femtech, quantum sciences, and women's brain health.

➀ Cambrian Era has been striving to become the Chinese version of NVIDIA for nine years, with investors already enjoying the stock price surge;

➁ Yuanhe Yuandian, an early investor, was confident in Cambrian Era's potential due to its technology and market prospects;

➂ Huawei's support accelerated Cambrian Era's commercialization, but the company faced challenges after losing Huawei as a major customer;

➃ Cambrian Era's revenue mainly comes from government-led projects, raising concerns about its sustainability;

➄ Despite challenges,公募基金 and ETFs have been investing in Cambrian Era, driven by the AI sector's growth and domestic substitution trends.

➀ Cerebras Systems' wafer-scale AI chip (WSE-3) outperforms the fastest GPUs by 57 times in executing DeepSeek-R1 models with 70 billion parameters.

➁ Cerebras CEO Andrew Feldman states that enterprise clients are highly enthusiastic about DeepSeek's new R1 inference model, with a surge in demand within ten days of its launch.

➂ The WSE-3 chip, made on a 12-inch wafer with TSMC's 5nm process, has 4 trillion transistors, 900,000 AI cores, 44GB on-chip SRAM, and a total memory bandwidth of 21PB/s, with a peak performance of 125 FP16 PetaFLOPS.

➃ DeepSeek-R1 offers performance comparable to advanced inference models from OpenAI at a low training cost and has been open-sourced, allowing tech firms to build AI applications and chip manufacturers to optimize for the model.

➄ Andrew Feldman emphasizes that while DeepSeek poses some risks, users should exercise basic judgment, as seen with the use of electric saws.

➀ DeepSeek仅花费600万美元用于计算能力以训练其R1模型，展示了使用更少或较不先进的芯片也能获得高质量的人工智能结果。

➁ 根据聊天机器人提供商DocsBot的数据，使用DeepSeek-R1上传和下载1百万token的成本分别为55美分和2.19美元。

➂ 深度学习模型R1在多个行业基准测试中与竞争对手的表现相当，甚至略胜一筹。相比之下，其他美国公司如OpenAI和xAI的模型虽然具有更大的输出窗口，但在性能测试中表现较差且成本更高。

➀ Codasip GmbH and RED Semiconductor International Ltd have signed a Memorandum of Understanding (MoU) to collaborate on developing advanced AI acceleration technologies.

➁ Under the MoU, RED will use Codasip Studio processor design tools to integrate its VISC technology with Codasip RISC-V cores.

➂ The collaboration aims to enhance AI processing capabilities through specialized accelerators.

➀ The article discusses the current and future impact of IoT, AI, and ML on various industries, including agriculture, smart homes, manufacturing, e-commerce, and automotive.

➁ It highlights the growing adoption of AI and ML in India, driven by government initiatives and academic support, with projections indicating a market value exceeding $15 billion.

➂ The text also outlines the technological challenges and the role of government programs in fostering innovation and development in these fields.

➀ Ambarella has expanded its N1 Edge GenAI family with the introduction of the N1-655 edge GenAI system-on-chip (SoC). This chip can decode up to 12 simultaneous 1080p30 video streams while processing the video and running multiple vision-language models (VLMs) alongside CNNs.

➁ The N1-655 supports on-device GenAI processing, making it suitable for applications such as AI boxes, autonomous mobile robots (AMRs), and security video recorders. It consumes less power compared to cloud processors, reducing operational costs and enhancing privacy.

➂ The expansion of Ambarella's edge GenAI SoC lineup aims to meet the increasing demand for on-premise AI processing, driven by the costs associated with cloud-based processing, including bandwidth, service fees, and concerns over data security and privacy.

➀ Electronic Design's Editor Bill Wong comments on the announcement of DeepSeek.

➁ The challenges faced by DeepSeek in the field of machine-learning investment are discussed.

➂ Insights into the future direction of DeepSeek's development are provided.

➀ Former Intel CEO Pat Gelsinger argues that the market's reaction to DeepSeek's breakthrough in AI is wrong.

➁ Gelsinger believes that reducing the cost of AI will expand the market, not contract it.

➂ He also emphasizes the importance of openness in fostering innovation and collaboration in AI.

➀ According to a survey by the European Council on Foreign Relations, populations of the founding BRICS nations have a more positive outlook on Trump's second term in office compared to those in Europe, Korea, and the UK.

➁ In India, over 80% of respondents believed that Trump's re-election would be beneficial for their country.

➂ The survey suggests that many respondents believe Trump could bring peace or reduce tensions in Ukraine, the Middle East, and in US-China relations.

➀ DeepSeek, a Chinese AI software developer, claims it can train leading AI programs at a fraction of the cost compared to top US developers.

➁ DeepSeek recently launched R1, a program trained on 2000 Nvidia GPUs, which outperforms models from OpenAI, Anthropic, and xAI's Grok.

➂ The company also offers V3, which is free but requires payment for integration with its infrastructure. V3 excels in inference speed and competes with the most advanced closed-source models globally.

➀ Microchip announces new AI chip advancements;

➁ Semiconductor industry sees growth despite global challenges;

➂ New cooling solutions from DeepCool and Noctua gain attention.

➀ Two research projects, KI-AmaR and XRai Vision, are receiving funding to develop AI-assisted systems for complex repairs and the use of XR technologies in industrial processes.

➁ The KI-AmaR project aims to enable complex repairs without specialized training using a mobile smart factory equipped with a 3D printer.

➂ The XRai Vision project focuses on supporting companies in adopting Extended Reality (XR) technologies like AR and VR to improve their workflows and productivity.

➀ On January 22, 2025, President Donald Trump announced a $500 billion investment in AI infrastructure at the White House.

➁ This project, called The Stargate Project, aims to build new AI infrastructure in the United States over the next four years, led by SoftBank and OpenAI.

➂ The initiative is expected to secure American leadership in AI, create jobs, and enhance national security.

➀ Qualcomm Technologies has unveiled the Snapdragon X, a new SoC designed for AI-enabled laptops and compact desktops, priced starting at $600.

➁ The Snapdragon X boasts advanced CPU and GPU capabilities, along with dedicated AI acceleration, aiming to make powerful computing more accessible.

➂ It is expected to power Windows 11 laptops and mini desktops from manufacturers like Acer, Asus, Dell Technologies, HP, and Lenovo.

➀ Numem, a Sunnyvale startup, has developed MRAM-based chiplets designed for AI systems. Sampling is expected to start in late Q4'25.

➁ The chiplets offer up to 4TB/s memory throughput and support 4GB per stack package, making them scalable for demanding AI workloads.

➂ Numem's technology claims to significantly reduce standby power and deliver up to 4x faster performance than HBM while maintaining ultra-low power consumption.

➀ Tom's Hardware reviewed the RayNeo Air 2s, a pair of augmented reality glasses from TCL's RayNeo subsidiary.

➁ The Air 2s offers attractive design and wide compatibility but lacks advanced features and has mediocre audio quality.

➂ Despite its drawbacks, the Air 2s is priced competitively, making it a good value for those seeking AR glasses.