2025-06-10 01:55Summary of Conference Call Records Industry Overview - The AI industry can be divided into hardware (computing power) and software applications, with the US market showing strong performance in hardware while software remains speculative, similar to the current state in the A-share market [1][2] - The global AI industry has seen significant advancements recently, with major model companies launching new deep reasoning models, although the A-share market has not yet reacted to these developments [2] Key Company Insights Broadcom - Broadcom's latest financial report showed adjusted net revenue of $15 billion, a year-on-year increase of 20%, slightly exceeding expectations. AI-related semiconductor revenue grew by 46% year-on-year [4] - The company expects next quarter's revenue to reach $15.81 billion, with AI-related revenue projected to grow by 60% [4] - Broadcom's new TH6 switch chip, with a total capacity of 102.4 TBBPS, is expected to accelerate the upgrade of Ethernet networking from 800G to 1.6T, enhancing the performance and reducing costs associated with GPU interconnections [3][4] Cradle - Cradle's financial performance exceeded expectations, with Q4 2025 revenue of $170 million, a year-on-year increase of 180% and a quarter-on-quarter increase of 26% [5] - The company anticipates next quarter's revenue to be between $185 million and $199.5 million, driven by strong demand from major clients like Amazon, which accounts for 61% of its revenue [5] Nvidia - Nvidia's stock has seen a recovery due to increased demand for AI infrastructure, with significant investments in AI from companies like Oracle and Meta [6] - Nvidia's collaboration with Q has resulted in a low cost of GPU cloud services, making it competitive against AWS and Google [7] - The company is maintaining its market position by investing in GPU cloud vendors and offering competitive pricing for its products [8] Market Trends - The AI industry is viewed as the next major emerging trend following the internet and mobile internet, with a strong emphasis on both hardware and software applications [2] - The North American market for computing power is expected to maintain high growth, with significant increases in light module shipments anticipated for next year [4] - The overall sentiment in the AI hardware market is optimistic, with recommendations to focus on leading North American companies and related sectors [4][8] Additional Insights - The introduction of new deep reasoning models by leading global companies is creating anticipation in the market, with expectations for rapid advancements in AI capabilities [2] - The shift towards 1.6T Ethernet technology is expected to significantly impact the supply chain and market dynamics in the AI hardware sector [3][4] - The competitive landscape is evolving, with companies like Deep Seek and others in the A-share market being closely monitored for their potential to develop next-generation AI models [2]