Chip stocks fall as Nvidia supplier highlights market uncertainty
Chip stocks fell across the board early Thursday after supplier NVDA made comments on its earnings call that reflected uncertainty about semiconductor demand this year.
South Korea-based SK Hynix ( 000660.KS ) makes memory chips for Nvidia GPUs (graphics processing units) — chips used in data centers to power artificial intelligence software. After reporting fourth-quarter earnings that beat analysts’ expectations, SK Hynix Chief Financial Officer Woo-Hyun Kim commented on the year ahead on a post-earnings call: “The 2025 memory demand outlook is crowded with inventory adjustments from PC and smartphone OEMs (original equipment manufacturers).” . ) and reinforces protectionist trade policies and geopolitical risks.
Following the comments, Navidi fell as much as 2 percent. British chip designer Arm ( ARM ) fell nearly 6%, and SK Hynix rival Micron ( MU ) fell about 4%.
Meanwhile, Navia’s rivals Advanced Micro Devices ( AMD ) and Broadcom ( AVGO ) both sank around 1%.
The stock’s decline followed yesterday’s rally following news of US President Donald Trump’s announcement of Stargate, a massive AI infrastructure project backed by OpenAI, SoftBank ( SFTBY ), Oracle ( ORCL ) and UAE-based MGX.
Kim’s comments highlighted the gap in the market for semiconductors used in consumer products and artificial intelligence in data centers.
In a Jan. 13 note to investors, Needham analysts said: “2024 sees a diversification of semiconductor companies’ exposure to different end markets. Semiconductor suppliers exposed to the PC, smartphone, industrial and automotive sectors have seen fundamentals under pressure over the year, with a combination of generally weak end demand conditions and excess inventory.
“Al Kfai, on the other hand, had a strong desire to accelerate the development of Al’s infrastructure.”
SK Hynix’s Kim also commented on this transition: “As demand for AI memory continues to grow, the memory industry is shifting from a commodity market driven by volume and cost to our custom market, which focuses on high-performance and high-quality products.
Kim added that demand for memory chips used in AI data centers is “expected to continue to be fueled by continued competitive investments by large technology companies in advanced AI training and reasoning capabilities.”
Still, Needham analysts warned in a note that the gap between AI and AI chip stocks could narrow in 2025 as growth in AI revenues slows. In recent quarterly earnings calls, big tech companies including Google and Microsoft have reported. High AI costs will begin to grow slowly in the future.