January 6, 2026 09:56 PM GMT Asia-Pacific Technology | Asia Pacific AI Supply Chain: CES implications, ASIC production, China AI chips Morgan Stanley does and seeks to do business with companies covered in Morgan Stanley Research. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of Morgan Stanley Research. Investors should consider Morgan Stanley Research as only a single factor in making their investment decision. For analyst certificati ...

Core Viewpoint - Hut 8 Corp. is considered undervalued despite its strong execution in colocation deals and potential for growth in AI and crypto sectors, with a price target raised to $55.00 from $30.00 while maintaining an "Outperform" rating [1][3]. Group 1: Company Positioning - Hut 8 is well-positioned to capture incremental Google TPU deployments, with over 2GW of potential expansion across its sites and pipelines [2]. - The company has 1.0GW net expansion optionality at River Bend and 1.05GW across its broader pipeline with Anthropic [3]. Group 2: Market Sentiment and Valuation - The stock is currently undervalued due to weak sentiment in AI and Bitcoin markets, rather than concerns about execution or lease quality [2][3]. - The Fluidstack lease valuation ranges from $4.49 billion to $7.12 billion, translating to a share price of $41.55 to $65.88 at 5%-7% cap rates, which exceeds the current market cap [3]. Group 3: Analyst Insights - Analysts highlight that improving colocation pricing is a positive indicator for crypto miners with ready-for-service capacity over 2026-27 [1]. - The price target increase reflects a 6% cap rate on the Fluidstack lease, a 25% haircut to the ABTC stake, and net debt considerations [3].

Core Insights - Groq has entered into a $20 billion technology licensing agreement with NVIDIA, which is not a legal acquisition but a non-exclusive technology licensing deal allowing NVIDIA to use Groq's hardware and architecture designs [2][3] - Groq's CEO Jonathan Ross and nearly all key members will join NVIDIA, while Groq will continue to operate as an independent company, retaining its core intellectual property [2] - The deal's significance extends beyond the monetary value, as it reflects NVIDIA's strategic positioning in the AI landscape amid regulatory challenges and market pressures [3][28] Group 1: Transaction Details - The transaction is valued at approximately $20 billion, which is enough to acquire GlobalFoundries entirely or represents a quarter of Intel's market value [3] - NVIDIA will acquire all of Groq's physical assets but not its intellectual property, indicating a focus on talent acquisition rather than a traditional acquisition [2] - Groq has raised a total of $1.8 billion in funding, with significant investments from entities like the Saudi sovereign wealth fund [5][27] Group 2: Groq's Technology and Market Position - Groq's core product, initially named TSP and later LPU, utilizes a unique architecture with 144-wide VLIW design, offering advantages in speed and efficiency [5][9] - The architecture's reliance on on-chip SRAM instead of external memory allows for fast access speeds but limits storage capacity, posing challenges for deploying larger AI models [6][7] - Groq's architecture is distinct from traditional ASICs and TPUs, focusing on deterministic system behavior and low latency, which are appealing for real-time inference scenarios [10][11] Group 3: Industry Context and Strategic Implications - The deal is seen as a strategic move by NVIDIA to solidify its position in the AI infrastructure market, especially in light of increasing regulatory scrutiny and competition [28][31] - The transaction may also serve as a means for NVIDIA to gain favor with U.S. and Middle Eastern stakeholders, potentially easing export restrictions on AI products [28][30] - The broader context includes a trend of large tech companies engaging in high-value agreements with promising startups to secure technology and talent without formal acquisitions [26][27]

[Table_Title] 研究报告 Research Report 26 Dec 2025 香港策略 Hong Kong Strategy 海通国际 2026 年 1 月金股 HTI January 2026 Top Picks 周林泓 Amber Zhou amber.lh.zhou@htisec.com [Table_yemei1] 观点聚焦 Investment Focus | [Table_summary] 行业 | 分析师 | 行业 | Top Pick | 看好理由 | | --- | --- | --- | --- | --- | | 组 | | | | | | | | | | Google 被 AI 搜索压制的估值已经在逐步释放：1） | | | | | | 通过 AI Overview 过渡 + AI mode 解锁交互式 AI 搜索 | | | | | | 功能，守护搜索市场，未来广告业务保持良好能见 | | | | | | 等大量 度；2）Google TPU 订单激增，Meta OpenAI | | | | | Alphabet (GOOGL US) | 订购 TPU V7 系列。预 ...

Core Viewpoint - Nvidia has agreed to pay approximately $20 billion for the technology licensing of Groq, a startup aiming to challenge Nvidia's dominance in AI application chips, specifically inference computing chips [1][5]. Group 1: Transaction Details - The deal involves a non-exclusive licensing agreement, allowing Nvidia to design server chips that could be cheaper and faster for running AI applications compared to its existing product line [2]. - Groq's valuation in this deal is about three times higher than its previous funding round valuation of $6.9 billion [1]. - Nvidia plans to integrate Groq's low-latency processors into its AI factory architecture, expanding its platform for broader AI inference and real-time workloads [3]. Group 2: Groq's Background and Performance - Groq was founded in 2016 by Jonathan Ross, who was involved in the early development of Google's AI chips, and has recently launched a cloud business allowing small developers to run open-source AI models [3]. - Groq has raised approximately $1.8 billion from investors, including Blackrock and Tiger Global Management, and has adjusted its revenue forecast downwards due to challenges in competing with Nvidia [6][7]. - The company had projected over $40 million in revenue from its cloud business this year, with overall sales expected to exceed $500 million [8]. Group 3: Market Context and Competition - Nvidia's chips are widely regarded as the most powerful and efficient solutions for developing new AI models, but there is a growing demand for lower-cost alternatives like Groq's chips [2][10]. - Despite Groq's advancements, Nvidia maintains a stronghold in the high-end AI chip market, with its chips being the preferred choice for major cloud service providers [7]. - Other startups are also struggling to challenge Nvidia, with many seeking acquisition opportunities, as seen in Intel's negotiations to acquire AI chip startup SambaNova [11].

证券研究报告 强于大市 投资要点： 一年内行业指数与沪深 300 指数对比走势： 资料来源：聚源数据，爱建证券研究所 相关研究 《人工智能月度跟踪：摩尔线程、沐曦股份 IPO 首发成功》2025-12-19 《电子行业周报：NVIDIA H200 芯片放松出口 限制》2025-12-16 《爱建电子专题报告：iPhone 折叠屏有望带来 产业发展拐点》2025-12-15 《电子行业周报：字节跳动发布豆包手机助手》 2025-12-08 《"朱雀三号"首飞在即，推动卫星通信行业 发展》2025-12-01 许亮 S0820525010002 0755-83562506 xuliang@ajzq.com 朱俊宇 S0820125040021 021-32229888-25520 zhujunyu@ajzq.com 行业研究 / 行业点评 2025 年 12 月 21 日 行业及产业 电子 TPU 需求上涨带动 Google 产业链发展 ——电子行业周报（2025/12/15-12/19） 本周（2025/12/15-12/19）本周 SW 电子行业指数（-3.28%），涨跌幅排名 31/31 位， 沪深 300 ...

Summary of J.P. Morgan's Semiconductor Research Call Industry Overview - **Industry Focus**: Semiconductors and Data Center Capital Expenditure (Capex) - **Key Companies Discussed**: NVIDIA (NVDA), Advanced Micro Devices (AMD), Broadcom (AVGO), Marvell Technology (MRVL), Micron Technology (MU) Key Points Data Center Capex Growth - Data center capital expenditure is projected to grow by **65%+ year-over-year (Y/Y) in 2025**, an increase from the previous estimate of **55%** [1] - This growth translates to an **incremental spend of over $115 billion** in 2025 compared to 2024, up from the earlier estimate of **$75-80 billion** [1] - The expected **incremental AI-related revenue** for NVDA, AMD, AVGO, and MRVL in 2025 is projected to be **$85 billion+** [1] 2026 Projections - Data center capex is now expected to track to **50%+ growth in 2026**, revised from **30%** [1] - This implies an **incremental spend of over $150 billion** in 2026, with potential upside to **$175 billion+** if demand continues to surge [1] - The strong demand for AI compute is anticipated to push 2026 data center capex growth into the **60%+ Y/Y range** [1] Company-Specific Insights - **NVIDIA (NVDA)**: Recently announced agreements with OpenAI and Anthropic contribute to a **$500 billion+ revenue backlog** through the end of 2026 [1] - **Broadcom (AVGO)**: Market misinterpretation of management's comments on a **$73 billion backlog** due to lead times [1] - **Marvell Technology (MRVL)**: Likely upside from AWS orders for Tranium3 [1] - **Advanced Micro Devices (AMD)**: Anticipated announcements regarding significant customer engagements in the next 6-9 months [1] Market Dynamics - The spending environment is expected to support a **40-50% compound annual growth rate (CAGR)** for the AI accelerator total addressable market (TAM) off a **$200 billion base in 2025** [4] - Strong AI server spending is projected to benefit companies like NVDA, AVGO, MRVL, AMD, and Micron [4] Future Visibility - While visibility into data center capex beyond 2026 is limited, current expectations suggest a **~10% Y/Y growth** in 2027 if spending remains flat [1] Additional Insights - The report emphasizes the importance of not solely focusing on traditional hyperscalers, as significant spending is also occurring with neoclouds and sovereign AI projects [4] - The analysis indicates that the semiconductor sector is well-positioned to capitalize on the ongoing AI demand, with several companies highlighted as top picks in the AI/cloud space [4] Conclusion - The semiconductor industry, particularly in the context of data center capex, is poised for significant growth driven by AI demand, with key players like NVDA, AMD, AVGO, and MRVL expected to benefit substantially from this trend [1][4]

You know, obviously there's there's been a volatile element that has come into the AI sentiment here. >> Absolutely. >> A lot of it is centered around Oracle and sort of funding of these projects, private credit and its involvement in all of this.Like what are your thoughts on this. Yeah, I mean we've definitely seen um you know some concern building you know kind of within the markets you know just in terms of the amount of capex that is being committed you know kind of Oracle and the open AI kind of side ...

It's time to spotlight the tech sector, look back at the past year in tech. Joining us now to help us do that is Daniel Newman, the CEO of Futurum. Daniel, thank you for taking the time to be with us today.You know, since we're talking tech, I guess I have to ask everyone's favorite or least favorite question, depending on how you look at it. Do you think we're in an AI bubble. >> Yeah, that's a great way to start the conversation.Good to speak with you. And what a year it has been. I am probably one of the ...

Speaking of AI today, a possible deal between OpenAI and Amazon might be valued north of10 billion. For that, we'll turn to Mackenzie Sagalas. Morning, Mac. >> Hey, good morning, Carl.This would be a big upgrade and the Amazon OpenAI relationship. We're talking about both a huge investment, plus a path for AWS to land OpenAI as a flagship customer, which is a real optics win for Amazon's Nvidia alternative. Now, as for what I'm hearing from a source close to the discussions, Amazon is looking at an equity s ...