Marvell Technology Targets $30 Billion Custom AI Chip Business as Hyperscalers Seek Control

Marvell Technology Raises Sales Outlook on Strong AI Demand

Marvell Technology shares rallied 6% on Tuesday after the chip designer unveiled an ambitious long-term sales outlook driven by surging AI demand from data centers1

1

. At its annual investor day, the Nvidia and Alphabet partner announced it now expects to generate approximately $20 billion in sales next year, up from a previous forecast of $18 billion1

1

. More dramatically, Marvell's sales outlook for fiscal 2031 jumped to between $70 billion and $90 billion, substantially exceeding the Visible Alpha consensus projection of $46.85 billion1

1

. The stock surge pushed Marvell among the top performers in the S&P 500 and Nasdaq on a day when both indexes reached record highs, with shares more than tripling in value since the start of the year1

1

.

Custom AI Chip Business Emerges as $30 Billion Opportunity

Marvell's AI chip opportunity centers on a strategic shift among hyperscalers who increasingly want control over the custom silicon powering their AI infrastructure2

2

. The company revealed it is already shipping custom silicon to all four major U.S. hyperscalers, with its custom AI chip business generating about $1.5 billion last year2

2

. Marvell now expects revenue from this custom AI chip business to exceed $12 billion in fiscal 2029, up from a previous target of more than $10 billion, before reaching approximately $30 billion by fiscal 20312

2

. CEO Matt Murphy described the progression as allowing hyperscalers to start with Marvell's standard products, move toward customized systems, and eventually commission fully custom silicon incorporating their own intellectual property2

2

. Management emphasized that the forecast is based largely on programs and customer positions already in place rather than requiring entirely new wins2

2

.

Google Partnership Signals Broader Hyperscaler Strategy

The relationship with Google demonstrates why Marvell's AI chip opportunity has become strategically important for cloud providers seeking customized solutions2

2

. In August, Marvell announced a broad agreement to supply Google with custom chips covering products tied to Google's TPU ecosystem, including inference accelerators, storage and networking controllers, and memory interfaces1

1

2

2

. The agreement could generate up to roughly $120 billion in revenue through fiscal 2033 if performance targets are met2

2

. Marvell's custom chief Will Chu stated the company has momentum across all tier-one hyperscalers and expects higher attach rates as AI systems become more complex2

2

. The opportunity is becoming more balanced between XPU products and what Marvell calls "XPU attach"—networking, memory, storage, and other components that connect to AI accelerators2

2

.

Strategic Partnership with Nvidia Creates Complementary Ecosystem

Marvell's rise in the custom silicon market does not necessarily position it as a direct Nvidia competitor2

2

. The companies announced a strategic partnership in March under Nvidia's NVLink Fusion platform, with Nvidia investing $2 billion in Marvell Technology2

2

. Under this arrangement, Marvell can provide custom XPUs and networking compatible with Nvidia's architecture, allowing customers to customize parts of their systems while retaining Nvidia technology elsewhere2

2

. This makes Marvell's AI chip opportunity more nuanced than a simple competitive scenario—the company can benefit from hyperscalers wanting proprietary silicon while remaining connected to the broader Nvidia ecosystem2

2

. The key question for investors is whether the shift toward customized AI infrastructure driven by strong AI demand in data centers can transform today's design wins into Marvell's projected $30 billion custom AI chip business, while for Nvidia investors, the signal lies in how much future AI spending remains concentrated in GPUs versus migrating toward custom silicon and connectivity layers2

2

.