3 Semiconductor Stocks to Buy Following the Recent Chip Sector Sell-Off

The market for artificial intelligence (AI) infrastructure remains hot, but on Wall Street, nothing can go straight up forever. Last week, semiconductor stocks as a class got caught up in a sell-off triggered by the lackluster guidance that Broadcom (NASDAQ: AVGO) delivered with its quarterly report on June 3, and many top names tumbled sharply.

Dips like these aren't unusual, and they can be great buying opportunities. Let's look at three chip stocks to buy now in the wake of this one.

Nvidia

When it comes to AI infrastructure, Nvidia (NASDAQ: NVDA) remains the dominant force with its graphics processing units (GPUs). The company has created a wide moat around AI model training with its CUDA software platform, which is where most foundational AI code has been written and optimized for its chips.

It is also well positioned to power the new workloads that are coming as a greater share of compute goes toward inference and agentic AI. The company smartly entered into an agreement with Groq for its language processing units (LPUs), which are chips based on static random-access memory designed specifically for inference, and incorporated them into its CUDA ecosystem.

This lets the company offer end-to-end servers optimized for inference, where Nvidia's GPUs handle one phase, and Groq's lightning-fast LPUs reduce latency. Nvidia has also developed its own ARM-based central processing units (CPUs) -- and CPUs will be needed in much greater proportions in data centers that are handling agentic AI workloads.

Nvidia has continued to put up extraordinary growth, with revenue surging 85% last quarter to $81.6 billion. The stock trades at a forward price-to-earnings ratio (P/E) of just 23 times this fiscal year's estimates and 16 times the consensus for its fiscal 2028, which ends in January 2028.

AMD

Advanced Micro Devices (NASDAQ: AMD) had been one of the market's hottest large-cap stocks this year, although it pulled back from its highs amid the chip sell-off. The company is riding two powerful waves with inference and agentic AI.

While AMD and its GPUs were largely left behind during the large language model training phase of the AI revolution, the company is much better positioned for the era of inference, which is less computationally intensive than training. And its ROCm software platform has greatly improved over the past few years. Its chiplet designs can pack in more memory, which is generally a more pressing need than raw computing power when handling inference workloads. AMD already has two $100 billion purchase commitments for its GPUs, which should be big growth drivers.

The company is also a leader in the data center CPU market, which will make it a prime beneficiary as the use of agentic AI expands. In data centers being built recently, the GPU-to-CPU ratio has been 8:1. For server clusters that are dedicated to powering AI agents, that ratio is expected to shift to 1:1. AMD management has said this is a $120 billion market opportunity in the coming years. AMD has strong technology in this area, and its new Venice architecture is designed for agentic AI.

Between its GPU and CPU opportunities, AMD is poised for skyrocketing growth in the coming years.

Image source: Getty Images

Broadcom

Another great stock to pick up after the recent chip sell-off is the one credited with setting it in motion: Broadcom. Investors were not happy that the company didn't raise its guidance for the next fiscal year's AI chip revenue, but this is still a company that is set to see explosive growth in the coming years. And now, buyers can grab the stock while it's trading at a forward P/E of just 20 times estimates for its fiscal 2027, which ends in June 2027.

Broadcom is a leader in data center networking technology, and the top designer of application-specific integrated circuits (ASICs) -- AI chips built to handle a narrow range of workloads more efficiently and cost-effectively than general-purpose GPUs. The company helped Alphabet develop its Tensor Processing Units (TPUs), a partnership that is a big growth driver by itself, and it is also benefiting as other hyperscalers come to it for help designing their own custom chips. Broadcom forecasts that this will be a more than $100 billion business for it in fiscal 2027, which is more than the $63.9 billion in total revenue it recorded last year.

The company is also a leader in optical networking and has been an early mover in co-packaged optics, which combine optical and electronic components within the same package to enhance bandwidth, reduce latency, and improve power efficiency in modern data centers.

While optical technology names have been all the rage in the markets lately, investors have largely forgotten Broadcom's strong position in this area, and are overlooking the ways that its optical interconnect technology directly ties into its ASIC business.

Geoffrey Seiler has positions in Advanced Micro Devices, Alphabet, and Broadcom. The Motley Fool has positions in and recommends Advanced Micro Devices, Alphabet, Broadcom, and Nvidia. The Motley Fool recommends Arm Holdings. The Motley Fool has a disclosure policy.