Qualcomm (NASDAQ:QCOM | QCOM Price Prediction) shares have seemingly been watching the semiconductor ascent from the sidelines in these past few years, gaining just over 9% in the past five years. Undoubtedly, so many investors are on the hunt for that catch-up trade that goes for a reasonable price of admission.
And while shares of Qualcomm have been a turbulent, less-rewarding ride for its shareholders, the name surely wasn’t spared from the latest barrage of turbulence hitting the semiconductor scene, with shares now down around 35% from its early-June peak — a moment when it seemed like Qualcomm stock would finally break out after going nowhere for so long.
Indeed, buyers of that late-Spring pop were punished quite severely. As the company looks to move on from the DRAM and NAND crisis weighing heavily on the smartphone market and the broader volatility hitting the semis, I do think the name might finally have the means to attract some of that AI excitement come the second half.
Qualcomm’s under pressure, but this might be more of an opportunity
Of course, the so-called RAM crisis isn’t going to magically solve itself in a few months. But, in my view, the firm has a slate of seriously impressive hardware that might spark a re-rating as investors look past the constraints facing smartphones for the on-device AI compute boom that likely lies on the other side.
Indeed, cloud AI compute has been all the rage, especially with hyperscalers pouring hundreds of billions to advance the cause. At the same time, we can’t forget about on-device AI, which stands to play a massive, growing role in driving down the costs of inference.
Apple (NASDAQ:AAPL) and its powerful A-series silicon and cutting-edge NPU (neural processing unit) in the latest and greatest iPhones might get most of the buzz. With Siri AI on the way and plenty of promise for Apple’s Foundation Models, especially the ones that run on device, edge AI could become a major theme in the fourth and final quarter of the year. And going into 2027, perhaps NPUs and edge AI hardware could be that AI catch-up trade that value investors were looking for.
Qualcomm’s Hexagon NPU looks like a game-changer
In any case, Qualcomm’s next-generation Hexagon NPU packs plenty of promise as Snapdragon looks to get serious about on-device AI. As language models shrink down to run on edge devices, questions linger as to whether NPUs could be the spark that drives a consumer device supercycle.
For Apple, I do believe it’s on the cusp of a massive AI-driven supercycle as the firm makes immense ground in the AI race with one big software update later this year. Of course, the RAM crisis represents a very tall hurdle to pass, but, in due time, I do think Apple and the rest will make it past, even if a violent crash in DRAM and NAND prices isn’t in the cards in the next 12 months.
In any case, Qualcomm’s NPU is impressive in its own right, with localized multi-billion parameter models possible on-device. That’s big, as local inference looks to take off in an effort to improve consumer AI economics.
While Qualcomm’s Hexagon catalyst won’t send shares surging overnight, I do think that over time that generative AI features are going to keep expanding and getting better. As more capability is shrunken into a smaller form factor, my guess is that the AI-driven hardware demand supercycle could span some number of years.
For Qualcomm, the Hexagon NPU architecture seems to check all the right boxes. Whether it’s power efficiency from the fused accelerator architecture, the zero-lag response from running things on device, offline support, or the “ambient” AI running behind the scenes to improve the little things that users don’t even notice, I do think that the upgrade cycle (or supercycle) rumored to hit won’t be limited to Apple.
The bottom line
As more investors lose interest in Qualcomm after a tough earnings report, I’d look to give the name a serious second look as its new NPUs enter the conversation while the stock goes for 15.5 times forward price-to-earnings (P/E), a multiple that makes no sense if you’re a believer in the imminent rise of AI inference on the edge.
Contact [email protected] for any questions or corrections.
Top Gaining Stocks
Top Losing Stocks