Q4 Tech Earnings Preview: The Great Divergence

Despite posting strong fourth-quarter earnings so far, the information technology sector has lagged the broader market this year. Valuation concerns, rising input costs, and doubts about AI's potential to generate a lasting return on investment have all weighed on performance. Tack on tariff threats and a market rotation out of mega-cap tech, and it's been a volatile start to 2026.

That's put the S&P 500 Information Technology Index's incredible three-year streak of hefty double-digit gains in jeopardy. As a result, the second half of Q4 earnings season will be critical for the info tech sector. With a packed slate of semiconductor, software, and computer memory companies set to report, investors will have no shortage of data points to track. However, the key theme to watch is an unusual one for the historically monolithic tech sector: performance divergence.

"Tech stocks have broadly underperformed in the new year, but there has been a lot of divergence underneath the surface as investors become more discerning around potential AI winners and losers," said Nathan Peterson, director of derivatives research and strategy at the Schwab Center for Financial Research. "While the AI-secular-growth story still brings the potential of investment gains, investors are acknowledging the potential disruption to some incumbent tech leaders and becoming increasingly selective with their AI investments."

Despite the early volatility seen so far this year, the AI infrastructure trade has remained in vogue, with investors still favoring companies that sell the "picks and shovels" of the AI era. The PHLX Semiconductor Sector Index, for example, has touched new all-time highs this year. But the move has been driven by fresh faces. Semiconductor equipment and computer memory makers are now leading the pack amid a surge in semiconductor memory prices, while prior leaders like Nvidia (NVDA) and Broadcom (AVGO) have fallen behind.

The sell-off in software-related names has continued this year as well. Shares of once high-flying software giants like Salesforce (CRM), Adobe (ADBE), and ServiceNow (NOW) have all been pressured. "Fears of potential AI encroachment into the software space have led investors to shoot first, rather than wait for earnings results to find out if the concerns are justified," Peterson noted.

For investors, the second half of Q4 earnings season will offer an opportunity to gauge AI infrastructure and software demand, as well as the sustainability of the historic rise in computer memory pricing—trends that will undoubtedly be critical to the info tech sector's performance this year and beyond.

Before discussing what to watch in the second half of Q4 earnings season, it's important to note that tech companies are off to a strong start. As of January 23, when 13% of S&P 500® companies had reported results, information technology companies posted 26.2% year-over-year earnings growth, according to FactSet's Earnings Insight.

Multiple tech sector leaders, including critical AI bellwethers, have shown earnings momentum already this quarter. Taiwan Semiconductor Manufacturing (TSMC), for example, reported earnings and revenue figures that were well ahead of analysts' consensus estimates. The world's largest semiconductor foundry also guided for 40% revenue growth in the first quarter and boosted its full-year 2026 capital expenditures guidance to a range between $52 and $56 billion. This was widely viewed as a sign of confidence in demand for advanced chips and the broader AI infrastructure buildout. 

Intel (INTC)—which has surged since August 2025 due in part to an $8.9 billion investment from the federal government—was a different story. The chip and CPU maker turned in disappointing guidance in its Q4 earnings report, leading to a sharp sell-off in the stock. But importantly, executives said that weak demand wasn't the issue. Instead, INTC simply didn't produce enough chips and CPUs to meet targets. "We're working aggressively to grow supply to meet strong customer demand…and fully capitalize on the vast opportunity AI presents across all of our businesses," INTC CEO Lip-Bu Tan said in a follow-up earnings statement.

While tech earnings have largely impressed thus far, lofty valuations in the sector—and equally lofty expectations—have led to underwhelming stock price performances, relatively speaking. Whether that continues will likely depend on the earnings results of some key players in the coming months.

Looking ahead, magnificent seven earnings will continue to draw investors' attention, with Nvidia, Alphabet (GOOGL), and Amazon (AMZN) still on deck. It will be critical to monitor whether these hyperscalers are committed to maintaining their aggressive AI infrastructure spending. However, investors may want to broaden their focus beyond mega-cap tech given the divergence in tech stock performance of late.

For example, a string of semiconductor earnings reports from Advanced Micro Devices (AMD), Broadcom, Qualcomm (QCOM), and more will provide investors with useful barometers of the AI infrastructure buildout and potentially insights into hyperscalers' ability to maintain their record capital spending.

Meanwhile, multiple software giants, including Salesforce and Snowflake (SNOW), may shed light on demand, capital spending, and margin trends in their struggling industry.

AI-related chip demand has been robust in recent years, and Wall Street largely expects that to continue heading into the second half of Q4 earnings season. RBC Capital Markets analysts, for example, estimated in January that annual semiconductor revenue derived from AI applications will surge to $550 billion by 2028, up from $220 billion last year.

Early tech earnings reports have offered some support for this outlook. As previously mentioned, commentary and spending trends from TSMC—and even INTC, which continues to face challenges—suggest chip and semiconductor equipment makers are preparing for increasing demand tied to the AI infrastructure buildout in 2026 and beyond.

With many chipmaker valuations already reflecting this bullish narrative, investors need to focus on whether results and guidance match expectations. Here are a few other key factors to watch:

• Inventory levels. Pay attention to chipmakers' inventory levels and inventory forecasts. Rising inventory can be evidence of weaker demand or more cautious customers, while falling inventory levels can suggest resilient demand that may support stronger margins.
• Capital spending trends. Monitor chipmakers' AI-related capital spending. Capital spending trends can potentially signal confidence—or a lack thereof—in future AI and data-center-related chip demand.
• Customer concentration. Watch for signs that chipmakers are attempting to diversify their customer bases. Heavy exposure to only a handful of hyperscalers has left some chipmakers vulnerable to changes in mega-cap AI spending plans.

Memory and storage stocks outperformed last year and have continued to be on fire in 2026. SanDisk (SNDK), Micron Technology (MU), Western Digital (WDC), and Seagate (STX) have skyrocketed in recent months along with chip equipment makers like Applied Materials (AMAT), Lam Research (LRCX), and KLA Corp. (KLAC). The cohort is benefitting from surging semiconductor memory prices, specifically DRAM and NAND prices. The AI infrastructure buildout has led to a combination of insatiable demand and strained supply for this type of high-bandwidth memory (HBM), which is essential for high-performance AI chips. However, the good times are unlikely to last forever.

"Historically, DRAM and NAND prices have been highly cyclical and subject to large price swings, both higher and lower," said Peterson. "It's difficult to predict how long the recent memory price surge will persist, but right now, there doesn't appear to be any signs of an inflection point."

Peterson believes computer memory stocks—and the key chip equipment stocks that have also benefited from rising semiconductor memory prices—will continue to experience higher volatility moving forward as investors gauge whether the memory pricing boom will continue or fizzle. Lofty valuations and earnings expectations will also pressure these firms to perform in coming quarters.

Still, comments from multiple tech CEOs have recently provided memory stock investors with hope the industry can continue its meteoric run. Sassine Ghazi, CEO of the semiconductor designer Synopsys (SNPS), told CNBC in late January that he believes the chip "crunch" will continue through 2027. "It's a golden time for the memory companies," he said.

Meanwhile, Winston Cheng, CFO of the PC maker Lenovo (LNVGY), said in an interview at the World Economic Forum in Davos, Switzerland that he expects memory prices to continue rising, calling the AI-infrastructure buildout a "very long" process. MU's Vice President of Operations Manish Bhatia also believes the memory supply crunch will last beyond this year due to demand from AI accelerators. "The shortage we are seeing is really unprecedented," he told Bloomberg in late January.

While many major players in the computer memory and chip equipment industries have already reported earnings, investors should closely monitor firms like MU and AMAT. Here's what to watch:

• Degree of beat. With memory and semiconductor equipment companies surging over the past year, beating Wall Street's earnings expectations might not be enough to boost these companies' stock prices. Consider tracking how much these firms can surpass estimates by this quarter relative to how much they were able to surpass estimates by in previous quarters.
• High-bandwidth memory pricing. Keep tabs on HBM inventory levels and production capacity. Supply and demand trends in this market will influence pricing and, by extension, near-term revenues and margins for memory makers.
• Forward guidance. Revenue and margin guidance may be more important for memory maker stocks than their quarterly results. Wall Street is looking for insights into how long the current AI-infrastructure-driven memory upgrade cycle will continue.

Nvidia CEO Jensen Huang famously made the claim that AI will "eat" software in a 2017 interview with the MIT Technology Review. The idea was that AI won't just fundamentally change software development as AI models take over the writing of code, but it may replace traditional software functions by automating tasks.

For years, many judged this idea to be somewhat farfetched, and software stocks continued to boom. But with agentic coding agents and a slew of other AI tools rolling out across the corporate world in recent years, there's been a tone shift among investors. Since the fall of last year in particular, tech investors have begun to fear Huang's prediction, and AI is now seen as a competitive threat that could eat into margins. As a result, the S&P North American Technology Software Index has plunged roughly 18% since its late September highs, with highly valued enterprise software names being hit particularly hard.

Over the long term, however, software companies may be able to adapt and further integrate AI into their products. Several Wall Street leaders have come out in defense of software names this year. Orlando Bravo, co-founder of the software investment firm Thoma Bravo, told CNBC at the World Economic Forum in Davos that he believes the narrative that AI will cannibalize the existing enterprise software ecosystem is "absolutely wrong."

Throughout the second half of Q4 earnings season, software investors will want to look for evidence to help gauge whether AI is truly a net negative for the industry. Here's what to watch:

• Margins and discounting. Look for signs of margin pressure due to companies offering discounts to close new deals or maintain customers. This could indicate waning demand for software due to increased AI-driven competition.
• Customer retention. Keep an eye on customer retention data or commentary. If customers are cutting back on their software investments or canceling services altogether, it could be a sign of a shift to AI-enabled alternatives.
• AI tool adoption. Watch for any insights into how much customers are using software companies' new AI features. If customers aren't adopting these tools, then the capital expenditures required to build them may not pay off.

Here's a breakdown of analysts' quarterly consensus earnings and revenues estimates for some major tech stocks reporting in the second half of Q4 earnings season:

• Alphabet (GOOGL): EPS $2.63 (up 22.4% year over year), revenue $111.4 billion (up 15.5% year over year)
• Advanced Micro Devices (AMD): EPS $1.32 (up 21.1% year over year), revenue $9.7 billion (up 26.2% year over year)
• Nvidia (NVDA): EPS $1.52 (up 70.8% year over year), revenue $65.6 billion (up 66.9% year over year)
• Oracle (ORCL): EPS $1.71 (up 16.1% year over year), revenue $16.9 billion (up 19.7% year over year)
• Broadcom (AVGO): EPS $2.0 (up 24.7% year over year), revenue $19.2 billion (up 28.5% year over year)
• Micron Technology (MU): EPS $8.48 (up 443.5% year over year), revenue $18.8 billion (up 133.7% year over year)
• Salesforce (CRM): EPS $3.04 (up 9.5% year over year), revenue $11.2 billion (up 11.8% year over year)

Not a complete list. Estimates are as of January 29, 2026, and are subject to change.

AI continues to fuel the largest tech infrastructure buildout in history. Hyperscalers are racing to secure the chips, computer memory, and semiconductor equipment needed to support AI services at scale. This scramble has led to immense opportunities for companies supplying that infrastructure—the pick-and-shovel sellers of this modern-day gold rush. Investors, understanding this, have flocked to these AI enablers, while punishing firms they fear may be disrupted by the rise of AI (like software firms).

Eventually, however, the tide may turn. The AI infrastructure buildout is unlikely to continue at its current pace forever, even if ongoing investment is necessary. For tech investors, the key this earnings season and beyond is to search for signs of an inflection point in tech infrastructure spending. If capital spending begins to slow after this current period of rapid expansion—just as it has during past technology buildouts like the early internet era—tech investors may need to reconsider how they're managing risk and allocating capital.