When corrections hit the stock market, crowds tend to form around the obvious: megacap blue chips, diversified business models, and brand moats. This playbook works, but I think it may already be priced into technology stocks.
The most lucrative opportunities in the Nasdaq Composite's (^IXIC +2.02%) recovery don't necessarily require identifying the safe stocks. It's finding which stocks are being punished for the wrong reasons but still stand to benefit from multi-year secular artificial intelligence (AI) tailwinds at a discount.
Image source: Getty Images.
A custom silicon maker nobody talks about
Nvidia has been the top name in the semiconductor arena for more than three years now. Few investors understand that hyperscalers like Alphabet, Amazon, and Microsoft are building their own custom AI chips to reduce GPU dependency on external suppliers.
Marvell Technology (MRVL +2.02%) is helping to enable this shift. The company rests at the intersection of two trends quietly supporting the AI infrastructure supercycle: custom ASIC design and optical interconnects. Marvell's data center revenue isn't correlated to AI spending in the most obvious way. Rather, the company's growth is tied to the direction of AI budgets, depending on required architectures as use cases and applications move toward deployment.
NASDAQ: MRVL
Key Data Points
Acceleration across AI infrastructure buildouts provides structural advantages to Marvell, regardless of which specific models and chip designs win the most market share.
AI memory is the new bottleneck
Micron Technology (MU +4.04%) is one of the most cyclical names in the AI conversation, which is exactly what makes it interesting right now. Historically, semiconductor memory has been treated like a commodity. However, expanding AI workloads are changing the entire memory and storage demand profile.
NASDAQ: MU
Key Data Points
High-bandwidth memory (HBM) has a finite supply base and a growing structural demand beyond its core consumer electronics applications. The market continues to perceive Micron's AI-adjacent revenue as a mere feature of its business rather than the backbone.
The cyclical label around Micron is compressing its valuation. Meanwhile, structural demand for the company's DRAM and NAND chips is building a durable floor.
Image source: Micron Technology.
The sneaky thing about Micron's upside is if the Nasdaq manages to bounce back by year's end, the company will likely recover more than its peers because it's been discounted twice already -- once for being cyclical and once for the macro drawdown.
A mispriced compounding platform
While Broadcom (AVGO 0.52%) isn't discussed the same way as Nvidia, the company arguably has the most durable AI profile of any major chip company. Broadcom's custom ASIC business is not speculative or imaginary. It's under a long-term contract with a growing number of hyperscalers, including Google's TPU program and OpenAI's accelerator ambitions.
NASDAQ: AVGO
Key Data Points
What makes Broadcom unique is that its data center revenue is layered on top of a networking and software business that generates cash flow regardless of AI cycle timing. Investors worrying about accelerating capital expenditures from big tech tend to discount Broadcom's exposure as structurally de-risked compared to pure-play AI names.
During a recovery, the market tends to rerate platform compounders more aggressively because they were indiscriminately sold off during the harshest periods of the correction. In my eyes, Broadcom fits this narrative perfectly.
