The AI Infrastructure Race That’s Squeezing Legacy Software Companies

A large conference room filled with international leaders at Davos 2026

IBM says the AI boom is squeezing software budgets, and Wall Street punished the company fast.

Quick Take

IBM said customers shifted late-June spending toward servers, storage, and memory.
The company linked that shift to delayed software deals and a weak second quarter.
IBM shares fell sharply after the warning, deepening market fear around enterprise software.
The damage was strongest in IBM’s mainframe and Transaction Processing business.

IBM Says Clients Put AI Hardware First

IBM CEO Arvind Krishna said clients moved their quarterly capital spending in the last weeks of June. He said they bought servers, storage, and memory first to secure supply before expected price increases. IBM said that shift hurt software budgets and delayed large deals. Reuters reported that IBM described the result as the clearest sign yet that the AI buildout is pressuring software spending across the sector.

The warning matters because it shows how quickly corporate money can move when AI gear is tight. Businesses are not just buying more software and smarter tools. They are also racing to lock in the hardware needed to run them. That kind of squeeze hits companies that sell legacy software, mainframe systems, and related services. It also helps explain why investors sold the stock so hard after the update.

Shares Fell As Investors Reacted To The Miss

IBM’s stock dropped about 26% in early trading after the company lowered its second-quarter outlook. Fortune India said the selloff spread to software and information technology services names even as the broader Nasdaq Composite traded higher. That reaction showed investors treating IBM’s warning as a sector problem, not just a company stumble. The market clearly saw the update as a sign that AI infrastructure spending is starting to crowd out other technology purchases.

Still, the size of the stock move can hide an important detail. IBM did not say software spending vanished. It said clients pulled money toward hardware tied to AI workloads. That is a different claim from saying software demand collapsed everywhere. The distinction matters for readers who want the real shape of the problem. It also matters for investors trying to judge whether the hit is temporary or the start of a longer shift.

IBM’s Own Results Show A Mixed Picture

IBM’s filing showed some business lines still growing. The company reported that software revenue rose 5% year over year in constant currency, while Red Hat grew 11% and Distributed Infrastructure rose 37% with a $500 million backlog. Those numbers show strength in parts of the software stack even as older areas weakened. IBM’s own guidance also kept full-year software growth outlook near 10%, which suggests management does not view the quarter as a permanent break in demand.

🇺🇸 IBM just suffered its worst day on record: a staggering 25% drop!

Second-quarter earnings fell short, and CEO Arvind Krishna is pointing fingers at a shift in spending from software to hardware.

Is this the sign of a bigger issue? 🤔

Clients panic-buying memory chips?!… pic.twitter.com/2xr497jhBG

— STAY JIT (@stayjustintime) July 15, 2026

That mixed picture fits a familiar pattern in tech. New waves of capital spending often hit older products first, while newer platforms keep growing. IBM said the shortfall was driven mainly by weaker Z systems and the related software stack, especially Transaction Processing. In plain terms, the pain was concentrated, not universal. For conservatives watching wasted spending and market hype, that is a useful reminder that big AI promises can still leave some firms scrambling for real cash flow.

What This Means For The Broader Tech Market

IBM’s warning fits a wider race for AI infrastructure that is pulling money into chips, servers, memory, and data centers. Research cited in the broader market discussion points to huge planned spending by major cloud companies and a growing gap between AI capex and near-term revenue. That backdrop helps explain why hardware vendors can win while software firms get squeezed. It also supports IBM’s core complaint that budgets are being redirected toward the physical buildout needed for AI.

At the same time, there is a competing explanation inside the numbers. Bloomberg commentary said rising memory prices and supply constraints are pushing customers to buy early, which means the shift may be driven partly by cost avoidance, not just AI enthusiasm. That does not erase IBM’s warning. It does show the pressure is more complicated than a simple “AI kills software” headline. For readers, the lesson is straightforward: the AI boom is reshaping budgets, and some legacy firms are getting hit first.