Key Points
- Alphabet’s recent unveiling of its memory compression technology has caused concern among investors in Micron, Sandisk, and Seagate; however, this apprehension overlooks a more significant narrative.
- Indeed, the availability of lower memory prices paired with more efficient, capable AI models could bolster the uptake of this technology, likely fueling a surge in memory demand.
- The current dip in stock prices for Sandisk, Seagate, and Micron Technology positions these stocks as compelling investment opportunities.
Alphabet‘s (NASDAQ: GOOG) (NASDAQ: GOOGL) made headlines last month by outlining its TurboQuant technology, intended to revolutionize the AI hardware landscape.
Essentially, TurboQuant serves as a compression technique that diminishes the size of large language models (LLMs) without compromising accuracy.
This innovation reduces the memory required for training LLMs. Google highlighted that TurboQuant aims to alleviate the surging memory costs attributed to a persistent shortage of memory chips.
Following Google’s announcement, shares of memory manufacturers—including Micron Technology (NASDAQ: MU), Sandisk (NASDAQ: SNDK), and Seagate Technology (NASDAQ: STX)—experienced substantial declines.
Investor fears surrounding the impressive revenue and earnings growth these companies have enjoyed—fueled by a favorable memory demand-supply equilibrium that elevates prices—are contingent upon Google’s algorithm.
However, a more nuanced examination suggests that Google may, in fact, have invigorated the prospects for the aforementioned stocks.
A closer analysis reveals why these three AI stocks stand to gain significantly from TurboQuant, positioning them as ideal choices for investors seeking to build substantial portfolios.
TurboQuant is set to enhance—not reduce—memory demand
The real-world implications of TurboQuant and its ability to diminish memory overhead in AI data centers remain to be seen. Yet, should this technology achieve success and widespread adoption—assuming Google opts for a broad rollout—it will likely augment memory demand.
This is due to the exponential growth in LLM sizes over recent years. For context, the largest LLM in 2019 contained a mere 0.09 billion parameters, a stark contrast to the astonishing 540 billion parameters recorded in 2022.
Parameters signify the numerical values that an LLM assimilates to process inputs and deliver responses. Thus, a model endowed with a higher parameter count theoretically possesses enhanced capabilities for comprehending inputs and generating accurate outputs.
Given this trend, the latest LLMs now boast over 1 trillion parameters, with several prominent models surpassing half a trillion. Alleviating a bottleneck, such as substantial memory demands, could empower AI companies to train larger and more proficient models.
Furthermore, Gartner recently noted that performing inference applications on a 1 trillion parameter LLM could cost 90% less by 2030, driven by reduced chip costs, increased chip utilization rates, and the transition to more economical chip options.
Consequently, the long-term trajectory of AI usage appears promising, buoyed by the advent of more cost-effective hardware.
Therefore, TurboQuant should not pose a threat to Micron, Sandisk, and Seagate; rather, the influx of affordable memory chips is poised to elevate their demand, as AI models require larger training datasets with an emphasis on superior accuracy and quality.
This reasoning substantiates buying these memory stocks, given that TurboQuant’s likely success will promote sustained demand for their products in the long run.
The memory supercycle can propel these stocks upward
Micron, Sandisk, and Seagate each operate within distinct sectors of the memory market. Micron specializes in compute-centric dynamic random-access memory (DRAM) and storage-focused NAND flash products, while Sandisk is purely dedicated to NAND flash. Seagate manufactures high-capacity hard-disk drives (HDDs).
Their offerings are integral to data centers, smartphones, personal computers (PCs), and any application necessitating data storage and computation.
AI has emerged as a pivotal driver for these firms, propelling demand for DRAM, NAND flash, and HDD storage crucial to data centers.
In fact, the memory demand stemming from data centers has been so robust that these companies are struggling to keep pace with chip production.
Industry analysts estimate that the ongoing memory shortage could extend until 2030. This projection is hardly surprising, considering Seagate’s assertion that AI-driven applications will generate a staggering 394 zettabytes (ZB) of data by 2028, up from 72 ZB in 2020, reflecting an impressive compound annual growth rate of nearly 24%. Technologies like TurboQuant may further accelerate data generation.
Given that generated data must be processed, Micron, Sandisk, and Seagate’s addressable markets are expected to flourish. Notably, all three companies have experienced robust revenue growth in recent years, driven largely by advancements in AI.
Analysts predict that their commendable earnings growth trajectory is set to continue.
Don’t be surprised if these companies sustain their impressive growth rates in the long run, as memory demand persistently escalates. Current valuations present an opportune moment to acquire these chip stocks while they remain attractively priced.
Seagate holds the highest valuation among the trio, trading at 24 times forward earnings—slightly above the tech-centric Nasdaq-100 index’s forward earnings multiple of 23.
Yet, Seagate merits this premium valuation, as its earnings are projected to surge by 61% and 53% in the current and subsequent fiscal years, respectively.
Micron stock, by contrast, boasts a forward earnings ratio of merely 6.3. Sandisk is also relatively inexpensive, with a forward price-to-earnings ratio of 14.
The compelling earnings growth opportunities for these companies indicate that their valuations are significantly below what they should ideally reflect.
This positions them for substantial stock price increases in the long run, meriting a spot in prospective million-dollar portfolios, particularly in light of their robust growth trajectories powered by significant investments in AI data centers and technological innovations poised to democratize AI.
Is now the right time to invest in Micron Technology?
Before committing to Micron Technology, consider the following:
The Motley Fool Stock Advisor analyst team recently identified what they consider the 10 best stocks for investors to consider, and Micron Technology wasn’t among them.
The selected stocks have the potential to deliver impressive returns in the forthcoming years.
Reflect on the inclusion of Netflix in this list on December 17, 2004… had you invested $1,000 at the time of our recommendation, it would be worth a staggering $550,348!* Similarly, Nvidia’s entry on April 15, 2005, could have turned a $1,000 stake into $1,127,467!*
Importantly, Stock Advisor’s total average return stands at an impressive 959%, significantly outperforming the S&P 500’s 191%. Don’t miss the latest top 10 list, accessible through Stock Advisor, and join a community crafted by individual investors, for individual investors.
Source link: Aol.com.
