Financial report outlook | "Dual Engine" Resonance of AI ASIC and Optical Interconnection May Consolidate Marvell Technology, Inc. (MRVL.US) Strong Gains in Q1 Financial Report.

Marvell Technology, Inc. (MRVL.US) will announce its first quarter performance for fiscal year 2027 after the US stock market closes on May 27 (Wednesday). As a leading company in the field of specialized AI chips (AI ASIC) and optical interconnects, the company's stock price has soared over 130% year-to-date, surpassing a market value of $170 billion, making it one of the top-performing companies in the Philadelphia Semiconductor Index. Market expectations are extremely optimistic. Currently, Wall Street is widely predicting that Marvell Technology, Inc.'s first quarter revenue will increase by 26% year-over-year to $2.4 billion, with adjusted earnings per share increasing by 21% year-over-year to $0.77. The data center business is expected to continue to be the main driver of the company's first quarter performance. The company expects first quarter data center revenue to increase by approximately 10% sequentially. Marvell Technology, Inc.'s previously announced fourth quarter performance for fiscal year 2026 showed a revenue increase of over 20% year-over-year to a record high of about $2.22 billion, with adjusted earnings per share of $0.80, surpassing market expectations. The data center business, closely related to AI training/inference super systems, contributed approximately $1.65 billion in revenue for Marvell Technology, Inc., accounting for about 74% of total revenue and achieving a year-over-year growth of about 21%. The company emphasized in its performance statement that the order volume for its data center business is growing at a "record pace." Market expectations for the upcoming financial report from Marvell Technology, Inc. are far beyond just meeting expectations. Goldman Sachs Group, Inc. analysts specifically point out that the most noteworthy items to watch for in the earnings call include the latest developments in potential cooperation with Alphabet Inc. Class C, updated performance guidance for fiscal years 2027-2028, and the impact of short-term fluctuations in gross profit margin. RBC Capital analysts also note that the strong trend in custom XPUs (accelerator processors), but tight wafer supplies may limit short-term upside potential. In the current situation where the stock price has significantly oversold optimistic expectations, the "value" of the first quarter financial report will directly determine whether Marvell Technology, Inc. can continue its upward trend. Performance Driver: Resonance of AI ASIC and Optical Interconnect "Dual Engine" AI ASIC is undoubtedly the first core pillar of Marvell Technology, Inc.'s valuation and fundamental growth. The global trend of generative AI has accelerated the development of AI chips by cloud computing and chip giants, who are competing to design the fastest and most efficient AI computing infrastructure clusters for advanced large AI data centers. With cloud computing giants like Alphabet Inc. Class C, Amazon.com, Inc., and Microsoft Corporation launching the "AI computing cost revolution" to accelerate the penetration scale of AI ASICs, the core competition in the inference era is no longer just "peak computing power," but rather the cost per token, power consumption, memory bandwidth utilization, interconnection efficiency, and the total cost of ownership after software and hardware coordination. In these core metrics, AI ASICs customized for specific workloads, compilers, and interconnects are naturally more cost-effective than general-purpose GPUs. Marvell's data center business achieved a record $6.1 billion in revenue for the 2026 fiscal year ending February; its custom chip business has an annual revenue of about $1.5 billion, with design orders from 18 cloud providers, building chips for customers like Amazon.com, Inc. (Trainium processor), Microsoft Corporation (Maia AI accelerator), Meta (new data processing unit), and handling the existing design work for Alphabet Inc. Class C's Axion ARM CPU. Last month, according to multiple media reports, Alphabet Inc. Class C is in talks with Marvell to jointly develop two new AI chipsone for memory processing units, intended to work in conjunction with Alphabet Inc. Class C's TPU, and another designed for AI inference tasks. Insiders revealed that the goal is to complete the design of the memory processing unit by 2027 and start trial production, with an initial plan by Alphabet Inc. Class C to produce nearly 2 million of these processors, complementing its production capacity of about 6 million TPUs annually. It is worth noting that Alphabet Inc. Class C is not looking to replace its existing partner Broadcom Inc., but rather introducing a third-party force into its custom chip supply chain to diversify risks. Following this, Amazon.com, Inc. announced on April 20 an expanded partnership with Anthropic. Anthropic has committed to investing over $100 billion in AWS technology over the next 10 years, and a key component of this agreement is the Trainium chip designed in collaboration with Marvell Technology, Inc. According to Wells Fargo & Company data, Anthropic has deployed over 1 million Trainium chips on AWS, and with the mass production of Trainium 3 in the second half of 2026, the corresponding computing capacity is expected to reach 2GW. Analysts at JPMorgan noted in an investment report, "Marvell Technology, Inc. is the long-term custom chip design partner for Amazon.com, Inc. in the Trainium project, currently accelerating the mass production of Trainium 3 and is expected to further expand market share in future Trainium 4." As large-scale cloud customers migrate from general-purpose GPUs to custom solutions, AI ASICs are becoming the certainty engine driving revenue growth for Marvell Technology, Inc. The market currently expects Marvell Technology, Inc.'s total revenue for fiscal year 2027 to grow by over 30%, reaching around $11 billion, with data center business expected to grow by about 40%. Optical interconnects are the second core pillar that Wall Street widely expects to drive Marvell Technology, Inc.'s stock price prospects and fundamental expansion logic, and also a key factor in NVIDIA Corporation's $2 billion investment in the company at the end of March. Once the connection distance exceeds about 10 meters, copper cable interconnects are no longer able to meet the bandwidth and distance requirements for large AI data centers, necessitating a shift to optical interconnect systems, where the core hardware in optical components is Marvell Technology, Inc.'s range of optical DSP and optical interconnect products. Optical interconnects focus on high-speed links between servers, switches, racks, and data centers. Looking further ahead, Marvell Technology, Inc.'s leading silicon photonics technology is also a key factor in the market's positive outlook for the company. Silicon photonics is not an alternative concept to optical interconnects, but rather a key technology that further advances optical connectivity from between cabinets, racks, and systems to within racks, systems, and even packaging. More precisely, silicon photonics technology focuses on advancing optical transmission towards the chip/package/I/O side, which is why Marvell Technology, Inc. chose to acquire Celestial at a high cost and announced the acquisition of Swiss high-speed optical chip company Polariton Technologies in April. A report issued by Oppenheimer on May 22 stated that Marvell Technology, Inc.'s optical interconnect business has had a compound annual growth rate of approximately 50% over the past five years, and is expected to exceed 60% growth this year. The company's management has increased its growth forecast for the 2027 fiscal year from 30% to 50% for the optical interconnect business. Additionally, analysts have become increasingly optimistic about the revenue prospects for Marvell Technology, Inc.'s SSD storage chip main control business. In large-scale training/inference systems, I/O bandwidth, persistent storage access efficiency, and memory pool interconnect efficiency are equally crucial for overall training costs and performance, and Marvell Technology, Inc.'s SSD main control chips, NVMe/CXL cache controllers, and high-bandwidth storage interconnect product lines are important components of the growing storage demand. While these highly specialized control ASICs may not be as prominent as the exponentially expanding AI ASIC business, they are essential for the data processing of large parameter AI models, directly driving the system efficiency and service quality at the data center level. Judging from recent statements from Wall Street, the market no longer sees Marvell Technology, Inc. solely as a general data center AI chip company, but is beginning to view the company as one that is not only accepting the wave of large cloud computing giants' self-developed XPU/ASICs, but is also leading the way in benefiting from the massive upgrades based on optical interconnects and the silicon photonics development dividend. Institutional target prices are raised intensively Ahead of Marvell Technology, Inc.'s latest financial report, several major banks have collectively raised their target prices for the stock. Citigroup maintains a "buy" rating for Marvell Technology, Inc. and has significantly raised its target price from $118 to $215, an increase of about 82%. Stifel also maintains a "buy" rating for Marvell Technology, Inc. with a target price raised from $140 to $210, and expects the company's first quarter performance to exceed market expectations and raise guidance. Bank of America Securities maintains a "buy" rating for Marvell Technology, Inc. with a target price raised from $125 to $200, and lists the stock as a "top pick." Wells Fargo & Company has raised its target price for Marvell Technology, Inc. from $135 to $195. Potential risks It is important to note that, in addition to optimistic expectations, investors in Marvell Technology, Inc. should be cautious of the following risks. First is the risk of high valuation. Marvell Technology, Inc. currently has a price-to-earnings ratio of about 64 times, significantly higher than the industry average, indicating that the market has already factored in a substantial level of "perfect expectations." If the management further raises the full-year performance outlook, the long-term growth logic of AI computing infrastructure will be further reinforced; however, if guidance indicates a slowdown in the growth of the data center business, the current high valuation will once again face market scrutiny. In addition, a slowdown in AI capital expenditure growth or deviations in company execution will also be a sharp edge hanging over Marvell Technology, Inc.'s high valuation. Second is the risk of execution and supply chain. Marvell Technology, Inc. has made consecutive acquisitions of multiple companies in the short term, including Celestial, Polariton, etc., and the integration risk should not be overlooked. The negotiations for cooperation with Alphabet Inc. Class C have "not yet formalized a contract," so there is uncertainty about whether the final agreement will materialize. At the same time, the tightness in advanced manufacturing capacity may constrain the speed of volume production for custom chip business. The company's management points out that advanced manufacturing capacity, advanced packaging, and substrate supply are still tight across the industry. Despite this, the company believes it has secured sufficient supply capabilities to support the strong growth prospects in the coming years. Additionally, there is the risk of customer concentration and competition. Amazon.com, Inc. is Marvell Technology, Inc.'s largest custom chip customer, and while the two have a five-year cooperation agreement, strategic adjustments by a single customer could have significant impacts. At the same time, Broadcom Inc. (AVGO.US) is aggressively expanding its annual revenue target for AI chips by over $100 billion, MediaTek has entered the TPU supply chain of Alphabet Inc. Class C, and competitors like AMD (AMD.US) and Astera Labs (ALAB.US) are adding pressure on Marvell Technology, Inc. In addition, the custom ASIC route itself is facing strong competition from NVIDIA Corporation's next-generation GPU platform if GPU cost-effectiveness continues to improve, some cloud customers may reduce the scale of ASIC deployments. Lastly, macroeconomic and geopolitical uncertainties are also important sources of pressure on Marvell Technology, Inc.'s short-term performance. Global trade tensions, evolving US chip export restrictions, and tariff policies will bring operational and demand-side risks, especially considering Marvell Technology, Inc.'s high dependence on large-scale cloud providers and the global supply chain.