By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
IBM's Q2 miss and subsequent 24% stock plunge reflect a shift in AI-related capex, with enterprises prioritizing hardware purchases due to shortages and anticipated price hikes. While this has led to a surge in IBM's distributed infrastructure unit, it also raises concerns about execution risk, potential margin compression, and the fragility of high-margin software and services revenue.
Risk: The risk that the capex shift is timing-based rather than demand-driven, leading to potential deal slippage or customer budget rationing, which could be recessionary.
Opportunity: The opportunity for IBM to accelerate hybrid cloud software uptake, particularly through its Red Hat and OpenShift offerings, as enterprises look to run private AI stacks.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
IBM (NYSE: IBM) didn't wait for its scheduled earnings date. On Tuesday, a week ahead of its July 22 report, the enterprise software and hardware giant released preliminary second-quarter results in a letter to investors from CEO Arvind Krishna. The numbers were disappointing. Revenue totaled $17.2 billion, up just 1% year over year and short of the company's own expectations.
Investors didn't take it well. Shares fell about 24% on Tuesday, one of the worst single-day drops in the company's history, and slid further on Wednesday to a 52-week low. IBM's market capitalization now sits below $200 billion.
Where to invest $1,000 right now? Our analyst team just revealed what they believe are the 10 best stocks to buy right now, when you join Stock Advisor. See the stocks »
But the most interesting part of the pre-announcement isn't the miss itself. It's Krishna's explanation of what happened in the last few weeks of June, because it says a lot about where technology budgets are actually going in the AI (artificial intelligence) buildout.
The shortfall was a sharp reversal. In the first quarter, IBM's revenue rose 9% year over year, led by infrastructure revenue that jumped 15% as the company's new z17 mainframe rolled out. IBM expected that mainframe momentum to fade as the launch wrapped up, guiding for infrastructure revenue to decline by a low-single-digit rate for the year.
Instead, second-quarter infrastructure revenue fell 7%, software grew just 5%, and consulting was flat. The deceleration reached the bottom line, too. Earnings per share of $2.27 declined 2% year over year, though earnings per share on a non-GAAP (adjusted) basis climbed 5%.
So, what happened? According to Krishna, IBM's clients abruptly changed their spending priorities.
"In the last few weeks of June, we saw clients shift their quarterly capex spend toward servers, storage, and memory purchases to secure supply constrained infrastructure ahead of expected price increases," Krishna explained in his letter. He added that the company "did not anticipate the magnitude of the capex reprioritization," and that numerous large deals failed to close on the timelines IBM expected, driving the majority of the shortfall. Krishna also noted that clients were distracted by rapidly evolving, industrywide cybersecurity concerns during the quarter.
Put another way, customers spent their quarterly technology budgets stockpiling hardware before prices went up, and other purchases got pushed out.
What stands out to me is that IBM's own report shows where those dollars landed. The company's distributed infrastructure business, which includes its Power servers and storage hardware, grew 37% year over year, its best performance in the company's reported history. The unit exited the quarter with a backlog of about $500 million. Even inside the company that missed, the money moved toward hardware.
The memory market shows the same scramble at a much larger scale. Micron Technology, one of the world's biggest memory-chip makers, reported revenue of $41.5 billion in its most recent quarter, up 346% year over year, as its DRAM selling prices more than doubled. And Micron says AI-driven demand for memory and storage has accelerated at a rate greater than the industry's ability to increase supply.
The AI buildout, in other words, is no longer just the giant cloud companies pouring capital into data centers. Ordinary enterprises are now competing for the same servers, storage, and memory, and they're pulling money from the rest of their technology budgets to secure it. That's a tailwind for memory and AI-infrastructure suppliers, whose products are the ones being stockpiled. And it's a new risk for any vendor whose quarter depends on large deals closing on schedule, because a customer racing to lock in hardware can put everything else on hold.
For IBM specifically, the July 22earnings callnow comes down to one thing: Were those slipped deals lost, or merely delayed? Management said it will discuss its full-year expectations on the call, and coming into this quarter, the company had guided for constant-currency revenue growth of more than 5% in 2026. If that outlook survives, most of this quarter's damage was a timing issue. If it comes down, the problem may be bigger than one quarter.
I think the bigger lesson, though, is the one Krishna spelled out himself. When customers are grabbing supply constrained hardware ahead of price hikes, the AI infrastructure cycle isn't cooling. It's strong enough to change the spending patterns of the world's largest companies -- and investors should expect it to show up in more earnings reports from here.
Before you buy stock in International Business Machines, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and International Business Machines wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $397,351! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,304,257!
Now, it’s worth noting Stock Advisor’s total average return is 934% — a market-crushing outperformance compared to 210% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
**Stock Advisor returns as of July 16, 2026. *
Daniel Sparks and his clients do not have positions in any of the stocks mentioned. The Motley Fool has positions in and recommends International Business Machines and Micron Technology. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
Four leading AI models discuss this article
"AI capex is real and reallocating budgets toward hardware, but IBM's miss reveals timing and execution risks that could pressure near-term multiples across the sector."
IBM's pre-announced Q2 miss (revenue +1% YoY at $17.2B vs expectations) and 24% stock plunge highlight classic AI capex diversion: enterprises reprioritized budgets toward servers, storage, and memory amid shortages and anticipated price hikes. IBM's distributed infrastructure unit surged 37% with a $500M backlog, while Micron's 346% revenue jump and doubled DRAM prices confirm AI-driven memory scarcity. This isn't cooling demand—it's accelerating, pulling dollars from software/consulting. Yet IBM's mainframe cycle fade, cybersecurity distractions, and slipped large deals signal execution risk. Forward P/E compression to ~13x may offer value if 2026 >5% growth guidance holds, but near-term volatility is elevated.
The article glosses over whether this hardware scramble is sustainable or a one-off pull-forward that leaves 2H demand exhausted; if enterprises have now overstocked ahead of tariffs or recession, Micron and broader AI-infra names could face sharp inventory corrections, while IBM's slipped software deals may prove permanently lost rather than merely delayed.
"The scramble for AI hardware is actively cannibalizing enterprise software and consulting budgets, creating a new, structural revenue volatility for legacy tech giants."
The market's visceral 24% reaction to IBM’s miss suggests investors are finally waking up to the 'cannibalization' risk in IT budgets. While the narrative focuses on hardware supply-chain constraints, the deeper issue is the volatility of enterprise software and consulting demand. If corporations are forced to reallocate fixed IT budgets toward 'must-have' AI infrastructure like Micron’s DRAM or IBM’s Power servers, high-margin software and services revenue becomes increasingly fragile. IBM’s 37% growth in distributed infrastructure is a bright spot, but it highlights a shift toward lower-margin, capital-intensive hardware that could compress long-term operating margins if software sales continue to stall.
The 'cannibalization' theory assumes fixed IT budgets; however, if the AI infrastructure buildout delivers measurable productivity gains, it could expand total IT spending rather than just shifting it.
"Customers front-loading hardware purchases to beat price increases is demand *acceleration*, not demand *strength*—it's a timing shift that creates a cliff risk in Q3 if capex budgets are truly constrained rather than merely rescheduled."
The article frames this as evidence of robust AI infrastructure demand, but IBM's 24% crash and guidance uncertainty suggest something more fragile. Yes, Micron's memory surge and IBM's distributed infrastructure +37% are real. But the article conflates *capex timing games* with *structural demand*. Customers front-loading purchases to beat price increases is classic cyclical behavior—it pulls forward demand, not creates it. The real risk: if memory prices stabilize or decline (Micron's DRAM pricing power is already showing signs of normalization in recent weeks), those stockpiled inventories become liabilities. IBM's slipped deals matter more than the article admits—if they're truly delayed rather than lost, we should see them close in Q3. If not, enterprises are rationing tech budgets, which is recessionary.
The article may be right that this is just timing. If IBM's full-year guidance holds and those $500M backlog deals close in Q3-Q4, the 24% drop is a buying opportunity in a company with 3.5% dividend yield and improving software margins.
"AI-driven hardware capex is a structural shift that could sustain demand for IBM's hardware and memory suppliers even if services growth remains weak, but the risk is timing and margin pressure if deals slip or supply tightness eases."
IBM’s pre-announced Q2 miss points to a real, but not fatal, shift in AI-related capex. The hardware backlog and 37% y/y growth in distributed infrastructure imply that buyers are stockpiling servers, storage and memory to hedge against shortages and price hikes, a theme Micron’s DRAM pricing and memory demand corroborate. The core risk is that the miss is timing, not demand destruction: if these deals slip or customers reallocate later, IBM could underperform. The bigger caveat: repeated capex cycles compress services margins and cloud software growth, and a broadly contractionary tech cycle could unwind this hardware bid. It remains a mixed picture, not a slam dunk.
Counterpoint: the 'AI infrastructure tailwind' claim could be a narrative; a single quarter of delayed deals doesn't prove durable demand. If enterprise purchasers pull forward as supply constraints ease or if cloud capex rebalances, hardware bets could fade and IBM's service mix disappoints again.
"Hardware stockpiling logically feeds hybrid cloud software acceleration at IBM, a linkage the panel missed."
Claude correctly flags capex timing vs structural demand, but nobody has flagged the second-order effect on IBM's Red Hat/OpenShift trajectory. If enterprises are hoarding hardware to run private AI stacks, that should accelerate hybrid cloud software uptake—not just slip deals. The $500M backlog may actually be early proof of embedded software demand that the article and panel have ignored.
"Hardware-first AI spending is creating a deployment bottleneck rather than an immediate software tailwind for IBM."
Grok, you're missing the friction of implementation. Enterprises aren't just 'hoarding' hardware for hybrid cloud; they are struggling with data gravity and legacy integration. The $500M backlog isn't proof of software uptake, but a symptom of hardware-first desperation. If IBM cannot bridge the gap between their distributed infrastructure and Red Hat's ecosystem, the software won't follow. This isn't an acceleration of the hybrid cloud; it's a bottleneck where hardware is bought but software deployment remains stalled.
"Hardware backlog growth doesn't confirm software demand acceleration; Q3 execution and deal closure timing will determine whether this is capex timing or demand destruction."
Gemini's 'implementation friction' framing is sharper than Grok's assumption that hardware hoarding automatically drives software adoption. But both miss the real test: IBM's Q3 earnings call. If management cites Red Hat integration delays as a reason for slipped deals, Gemini wins. If they cite customer indecision or budget freezes, Claude's recessionary thesis gains weight. The $500M backlog tells us nothing about software velocity—only that hardware moved. That's not nothing, but it's not proof of hybrid cloud acceleration either.
"Backlog alone won't translate into durable IBM revenue without effective OpenShift monetization through services and channel partners."
Challenging Gemini: implementation friction is real, but you’re assuming OpenShift monetization hinges on customers running AI on-prem. The bigger lever is IBM's partner and consulting network converting hardware capex into recurring software and managed services revenue. If OpenShift adoption stalls, the $500M backlog could vaporize. Watch channel commitments, training investments, and services margins; a hardware-led rush without durable software monetization risks depressing long-run profitability and leaving the capex binge hollow when demand normalizes.
IBM's Q2 miss and subsequent 24% stock plunge reflect a shift in AI-related capex, with enterprises prioritizing hardware purchases due to shortages and anticipated price hikes. While this has led to a surge in IBM's distributed infrastructure unit, it also raises concerns about execution risk, potential margin compression, and the fragility of high-margin software and services revenue.
The opportunity for IBM to accelerate hybrid cloud software uptake, particularly through its Red Hat and OpenShift offerings, as enterprises look to run private AI stacks.
The risk that the capex shift is timing-based rather than demand-driven, leading to potential deal slippage or customer budget rationing, which could be recessionary.