IBM's Mainframe Resilience: How AI Impacts Corporate Hardware Budgets

TL;DR

• IBM says its latest mainframe cycle remains strong, but recent quarter-to-quarter swings show that corporate capex timing can sharply affect sales when buyers redirect budgets toward AI hardware.
• The company’s CEO argues AI is not killing mainframes; instead, AI demand is reshaping infrastructure spending and can temporarily crowd out purchases of large systems like IBM Z.
• For enterprises, the near-term story is less about mainframes disappearing and more about how AI, servers, storage, and memory compete for the same hardware dollars.

Mainframe sales hit by a spending shift, not a structural collapse

IBM’s recent messaging suggests that its mainframe business is under pressure from timing effects in customer spending rather than from a permanent loss of relevance. In the latest reporting period highlighted by IBM, the company said it ended 2025 with momentum, citing strong adoption of its next-generation mainframe platform and more than $12.5 billion in generative AI business. At the same time, reporting in July described a disappointing quarter for IBM’s infrastructure segment, with mainframe sales weakening as customers redirected budgets toward AI-related hardware purchases.

The key distinction is that IBM is not saying the mainframe market has vanished. Instead, it is arguing that buyers are making tradeoffs in a constrained capital spending environment, especially as AI infrastructure demand intensifies.

Why AI is affecting hardware budgets

According to IBM CEO Arvind Krishna, some clients shifted quarterly capex in late June toward servers, storage, and memory to secure supply-constrained infrastructure ahead of expected price increases. That shift appears to have pulled spending away from mainframe purchases in the short term.

This matters because enterprise hardware budgets are not infinite. When companies race to assemble AI-ready infrastructure, they often prioritize components that support model training and deployment, which can temporarily reduce appetite for other large-ticket systems. IBM’s own explanation frames the issue as a reprioritization of spending, not a rejection of mainframes.

IBM still sees value in the mainframe

IBM’s broader results indicate the company continues to see its mainframe platform as strategically important. In January, IBM said its latest mainframe generation had driven robust adoption and helped cap a strong 2025, with the company entering 2026 with confidence in revenue growth and higher free cash flow. Other reporting has also pointed to strong annual revenue performance for IBM’s Z mainframe line in recent cycles.

That suggests the mainframe remains a durable product for workloads where reliability, security, transaction throughput, and large-scale data processing still matter. Even as the AI boom reshapes budgets, IBM appears to believe mainframes remain embedded in the long-term infrastructure plans of large enterprises.

The near-term challenge: timing versus momentum

The contrast between different quarters is important. One report described a strong quarter in which IBM’s mainframe and infrastructure revenue benefited from demand tied to AI deployment. Another, from July, described a weaker quarter where infrastructure revenue fell and IBM blamed customers’ late-quarter shift into AI infrastructure purchases.

That volatility suggests IBM’s mainframe business may be increasingly exposed to the rhythm of broader infrastructure cycles. When AI demand is intense, enterprise buyers may accelerate spending on systems they view as immediately necessary for AI operations, delaying other purchases until later budget cycles.

What this means for corporate hardware planning

For enterprise IT leaders, IBM’s situation reflects a broader budget tension: AI is not only a software initiative, but also a hardware one. Training and running AI workloads can require large investments in compute, storage, and memory, which can push organizations to reallocate spending across their infrastructure portfolios.

That does not automatically weaken mainframes over the long term. In some cases, AI growth may reinforce the value of mainframes if companies need trusted back-end systems to handle sensitive transactions, compliance-heavy workloads, and large data estates. But in the short term, AI can absolutely compete with mainframes for the same capital dollars.

What to watch next

The most important question is whether IBM’s recent mainframe softness is a one-quarter budget shift or part of a longer pattern. If AI infrastructure demand keeps pulling hardware spending forward, mainframe sales could remain uneven from quarter to quarter.

For now, IBM is drawing a clear line: AI is changing how corporations spend on hardware, but it is not, in IBM’s view, ending the mainframe era.

AndroGuider Team

Articles written by the AndroGuider team. We try to make them thorough and informational while being easy to read.