Introduction

Custom artificial‑intelligence (AI) chips are becoming a central driver of growth in the technology sector. Recent reports indicate that Amazon (AMZN) is exploring sales of its Trainium ASICs beyond internal use, suggesting that hyperscale data centers are expanding their appetite for purpose‑built silicon. This shift is fueling rapid revenue expansion for firms that specialize in designing and manufacturing AI‑optimized processors.

Broadcom’s Accelerating Momentum

Broadcom (AVGO) stands out as a leading beneficiary of this trend. In the second quarter of its fiscal year 2026, the company posted total revenue of $22.2 billion, a 48% increase year over year. Within that, AI chip sales surged to $10.8 billion, up an astonishing 143% compared with the same quarter last year. Adjusted earnings per share climbed 54% to $2.44, reflecting the high‑margin nature of AI hardware. Broadcom’s product portfolio includes high‑profile contracts with Alphabet, Meta Platforms, and OpenAI, positioning the firm as a trusted supplier for the most demanding workloads.

Looking ahead, Broadcom forecasted more than a 200% year‑over‑year increase in its AI chip segment for the following quarter, underscoring confidence in sustained demand. Although the stock fell 3.39% after the earnings release, the underlying growth trajectory suggests that the market may still be underestimating the long‑term upside.

Marvell’s Strong Growth Path

Marvell Technology (MRVL) is another key player capturing the AI chip surge. In the first quarter of its fiscal year 2027, the company reported record revenue of $2.4 billion, a 28% increase from the prior year, with adjusted earnings per share rising 29% to $0.80. Data‑center revenue now represents 76% of Marvell’s total sales, highlighting the strategic shift toward serving hyperscale customers.

Marvell expects its revenue to accelerate in each subsequent quarter of FY2027, reinforcing its commitment to meet rising AI chip orders. Although Broadcom’s AI chip revenue dwarfs Marvell’s—$10.8 billion versus $1.8 billion—the two firms together command roughly 95% of the custom AI ASIC market, demonstrating the concentrated nature of this high‑growth segment.

Implications for Investors

The convergence of expanding AI workloads, hyperscale cloud spending, and the development of specialized silicon creates a compelling investment narrative. Companies like Broadcom and Marvell, with deep design expertise and established relationships with leading cloud providers, are well positioned to capture ongoing capital expenditures. Investors seeking exposure to AI hardware can consider these stocks as part of a diversified technology portfolio.

Conclusion

Custom AI chip demand is accelerating, driven by cloud providers’ need for power‑efficient, high‑performance processing. Broadcom’s explosive revenue growth and Marvell’s solid earnings momentum illustrate how the market’s shift toward specialized silicon is translating into tangible financial results. As the AI hardware landscape continues to evolve, these two firms are likely to remain at the forefront, offering investors a clear avenue to benefit from the AI revolution.