How Does DeepSeek Impact Intel Stock?

China’s new AI model DeepSeek roiled the U.S. big tech stocks on Monday with semiconductor names taking the biggest hit. The industry expects the resource-light new model could usher in a wave of more efficient AI models, hurting demand for AI hardware. The worst hit were AI bellwether Nvidia (NASDAQ:NVDA) and Broadcom (NASDAQ:AVGO) which saw their stock prices decline by about 17% each. Memory maker Micron was down 12%. However, Intel stock (NASDAQ:INTC) held up pretty well declining by just about 2.5%. We think that Intel stock could benefit as DeepSeek shakes up the AI market. Here are a couple of reasons why.

Companies Free-Up GPU Capex, Spend On General Purpose Compute

Generative artificial intelligence has dominated the computing and semiconductor market narrative over the last two years. This has negatively impacted Intel, which primarily focuses on CPUs. Companies have prioritized investments in securing as many GPUs as possible, reducing demand for traditional CPU-based servers, largely driven by FOMO  or the “fear of missing out” on the compute capacity needed to deploy AI.  However, DeepSeek has reportedly restructured the foundation of AI models, emphasizing software-driven resource optimization over hardware dependency. This could translate into fewer GPUs to carry out tasks. In fact, the company claims that it spent just about $5.5 million to train its V3 model, compared to the $100s of millions that the likes of OpenAI are reported to spend.  DeepSeek’s model is open source, and it is likely that many big tech companies could likely take inspiration from its methods to cut down costs. If adopted widely, this could cool off the demand for AI computing power. Separately, if you want upside with a smoother ride than an individual stock, consider the High Quality portfolio, which has outperformed the S&P, and clocked >91% returns since inception.

Over the past two years, companies have likely under-invested in traditional CPU-based servers, eventually setting the stage for renewed spending in the general-purpose computing space. This could benefit Intel. Intel’s latest server chips, such as the Sierra Forest and Granite Rapids, are well-reviewed and utilize the advanced 3-nanometer “Intel 3” process node, which could aid the company’s growth.

Smaller AI Models Could Find Their Way Into PCs

DeepSeek is seen as a lighter and more efficient model and the methods the company has used here could potentially be deployed more widely into various applications.  Besides its full-scale 671-billion-parameter model, the company has also created smaller distilled versions that can run on relatively basic hardware such as PCs. The distilled version reportedly outperforms some larger models on key benchmarks for specialized tasks. This could enable businesses and institutions to create models that are optimized for their specific fields, such as financial modeling or healthcare. If there is a shift toward more lightweight models, it could potentially also play in Intel’s favor, as the company dominates the personal computing market – holding about 60% of the PC CPU market.

Intel’s Fledgling GPU Business, Foundry Can Benefit

Now Nvidia has dominated the AI accelerator market, with its GPUs which have been viewed as the best performing for AI training and inferencing with rival AMD coming in at a distance second place.  However, DeepSeek’s relatively lower computing power requirements could also open the door for other players to compete in the market. This benefits Intel in a couple of ways. For one, Intel is expanding its presence in the AI processor space with its Gaudi 2 and upcoming Gaudi 3 AI accelerators which are viewed as being more cost-effective compared to peers. If demand for high-performance GPUs potentially eases, Intel’s more affordable AI accelerators could become a good alternative for companies looking to reduce costs. Moreover, Intel’s foundry business could also see benefits. Customers such as Microsoft and Amazon have been contracting Intel to fabricate some of their of custom chips, including AI accelerators. This could gain momentum if these big tech companies use more of their own in-house chips instead of Nvidia’s high-end chips.

The decrease in INTC stock over the last 4-year period has been far from consistent, with annual returns being considerably more volatile than the S&P 500. Returns for the stock were 6% in 2021, -47% in 2022, 95% in 2023, and -60% in 2024. The Trefis High Quality (HQ) Portfolio, with a collection of 30 stocks, is considerably less volatile. And it has comfortably outperformed the S&P 500 over the last 4-year period.

Why is that? As a group, HQ Portfolio stocks provided better returns with less risk versus the benchmark index; less of a roller-coaster ride as evident in HQ Portfolio performance metrics. Given the current uncertain macroeconomic environment around rate cuts and multiple wars, could INTC face a similar situation as it did in 2021, 2022, and 2024 and underperform the S&P over the next 12 months – or will it see a recovery?

Intel stock trades at about $20 per share currently or just over 20x consensus 2025 earnings, which is reasonable in our view. We value Intel stock at about $27 per share, well ahead of the current market price. See our analysis of Intel’s valuation for a closer look at what’s driving our price estimate for Intel. Also see our Intel upside analysis on How Intel Stock Can Surge 3x To $60.  On the other hand, see our counter scenario which explores how Intel Stock Could Dive To $10.

[1] Returns as of 1/28/2025
[2] Cumulative total returns since the end of 2016

Invest with Trefis Market Beating Portfolios

See all Trefis Price Estimates