By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
The panel agrees that IBM's Q2 warning reflects a temporary shift in capex towards infrastructure, with software budgets expected to recover once infrastructure bottlenecks ease. However, there's disagreement on whether this is a cyclical or structural issue, and whether it poses a sector-wide risk or a company-specific challenge for IBM.
Risk: Prolonged crowding out of enterprise software budgets by hyperscalers' capex growth, or a permanent shift from OpEx to CapEx spending, leading to margin pressure for SaaS firms.
Opportunity: Software recovery once infrastructure investments stabilize and AI productivity gains materialize, potentially leading to a rebound in software investments and margins.
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 →
July 14 (Reuters) - IBM sparked a market rout on Tuesday after forecasting second-quarter revenue below estimates and signaling that businesses were favoring spending on data-center infrastructure over software, the starkest sign yet of AI's growing toll on the sector.
Shares of Big Blue slumped 20% in premarket trading, dragging other software stocks and Dow futures lower. The iShares Expanded Tech-Software Sector ETF was last down more than 4%.
Software investors have long been on edge over fears that AI tools capable of automating routine work could pose an existential threat to the industry. Tuesday's announcement showed that even the boom in spending on servers, chips and networking gear for AI was eating into software budgets.
"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," IBM CEO Arvind Krishna said in a letter to investors.
"While we anticipated some supply-chain related impact in our expectations, we did not anticipate the magnitude of the capex reprioritization," Krishna said, adding that the company had "faltered" in adapting quickly enough and that "numerous large deals" had failed to close as expected.
According to the preliminary results, the company expects revenue of $17.2 billion during the quarter, compared with analysts' estimate of $17.86 billion, according to data compiled by LSEG.
Adjusted earnings per share is expected to be $2.93, compared with the estimate of $3.02.
"This is an ugly moment for IBM and software stocks... the big question will be how long the shift to infrastructure and cybersecurity lasts," said Chris Beauchamp, chief market analyst at IG Group.
"A few more months might be bearable, but more than that and serious questions will be asked all over again about software stocks."
Microsoft, ServiceNow, Salesforce and Intuit fell between 3% and 5%.
(Reporting by Harshita Mary Varghese in Bengaluru; Editing by Shilpi Majumdar and Saumyadeb Chakrabarty)
Four leading AI models discuss this article
"IBM's miss signals near-term software budget pressure from AI infrastructure capex, but the effect is likely transitory rather than existential."
IBM's Q2 warning ($17.2B rev vs $17.86B est) and 20% premarket plunge highlight how AI-driven capex is cannibalizing software budgets, with clients prioritizing constrained servers/storage over licensing. This triggered a 3-5% drop in MSFT, NOW, CRM, INTU and a >4% hit to the iShares Expanded Tech-Software ETF. Krishna explicitly cited last-minute capex reprioritization and stalled large deals. While the article frames this as an existential threat to software, history shows infrastructure booms (cloud in 2010s) eventually lift adjacent software layers once foundational build-out stabilizes. Missing context: IBM's legacy mix may amplify the signal; hyperscalers' own software margins remain robust.
The shift may not be temporary; if enterprises conclude AI infrastructure spend yields faster measurable ROI than incremental software licenses, multi-quarter budget compression could become structural, forcing software valuations (often 10-15x forward revenue) to re-rate permanently lower.
"The market is conflating IBM’s specific failure to capture enterprise wallet share with a structural decline in the broader software-as-a-service industry."
IBM’s 20% drop is a classic 'capex cannibalization' event, but the market is overreacting by painting the entire software sector with the same brush. While IBM struggles with legacy integration and consulting-heavy revenue models, SaaS leaders like ServiceNow or Salesforce have higher switching costs and stickier recurring revenue. The real risk isn't that AI kills software budgets; it's that enterprises are prioritizing hardware 'plumbing' before they can build the 'applications' that run on top of it. We are in a transitional phase where infrastructure must be bought before software can be deployed at scale. This is a rotation, not a sector-wide obsolescence event.
If the 'AI infrastructure' phase lasts 18+ months, software companies will face a prolonged period of stagnant growth that their current high valuation multiples cannot justify.
"This is capex *timing* disruption, not demand destruction—but the article provides no forward guidance or pipeline data to confirm the distinction, leaving material uncertainty."
IBM's miss is real but narrower than the panic suggests. Revenue shortfall is $660M (3.7%) and EPS miss is $0.09 (3%)—material but not catastrophic. The actual signal: enterprise capex *reallocation*, not destruction. Clients aren't spending less overall; they're front-loading infrastructure before price increases. This is cyclical, not structural. Software will recover once infrastructure bottlenecks ease (likely Q4 2024 or Q1 2025). The real risk: if this signals *demand* weakness rather than timing, we'd see it in IBM's forward guidance and pipeline—the article doesn't detail those. Also: IBM's software segment (lower-margin, slower-growth) is disproportionately exposed; pure-play SaaS (CRM, analytics) may weather this better.
If enterprise IT budgets are genuinely constrained and capex reallocation reflects *total spend ceiling* rather than timing, then software compression could persist 2-3 quarters, forcing multiple compression across the sector that doesn't reverse quickly.
"Near-term software budgets may not shrink meaningfully due to AI, and IBM’s miss is likely idiosyncratic rather than indicative of a sector-wide trend."
IBM's miss looks more like a quarterly miscue than proof of a sector-wide AI squeeze. The capex shift toward servers and storage suggests customers are building AI capacity, not abandoning software investments. If supply chains normalize, hardware demand could stabilize while software tools tied to AI—security, analytics, and automation—stand to gain from higher productivity. IBM's weaker guide (Q2 revenue $17.2b vs $17.86b; adj. EPS $2.93 vs $3.02) and a 20% premarket drop may reflect company-specific delays and optics, not a universal drawdown in software budgets. Longer term, ROI from AI should sustain software demand even as hardware capex remains a factor.
The strongest counterpoint is that persistent AI-driven capex could indicate durable software softness ahead; IBM's guidance might reflect a broader pullback in software budgets, not just a one-off blip.
"Persistent hyperscaler capex growth risks multi-year software budget compression beyond IBM-specific issues."
Claude's timing optimism (recovery by Q1 2025) underestimates second-order effects: if hyperscalers continue 30-40% capex growth into 2025 to chase AI training scale, enterprise budgets face multi-year crowding out. IBM's legacy exposure doesn't fully explain the signal—pure SaaS firms like CRM and NOW also saw 3-5% sympathetic drops, hinting at broader pipeline scrutiny.
"The transition from OpEx-based software to CapEx-heavy AI infrastructure is forcing a structural re-evaluation of software ROI that will compress margins."
Grok is right to flag the 'crowding out' effect, but everyone is ignoring the CFO's role here. Enterprises are moving from 'AI experimentation' to 'AI production,' which requires a shift from OpEx-heavy software subscriptions to massive, front-loaded CapEx for private infrastructure. This isn't just a timing delay; it's a fundamental change in IT accounting. If CFOs force software spend to justify its own ROI against these massive hardware outlays, SaaS margins will face permanent downward pressure.
"The OpEx-to-CapEx shift is real, but its durability depends on whether total enterprise IT budgets are expanding, flat, or contracting—data the article doesn't provide."
Gemini's OpEx-to-CapEx accounting shift is the sharpest insight here, but it conflates two separate risks. Yes, CFOs are rebalancing budgets. But the real test: are enterprise *total IT budgets* growing, flat, or shrinking? If growing, software recovers once infrastructure stabilizes (Claude's case). If flat or shrinking, Gemini's permanent margin pressure thesis holds. Nobody's asked whether IBM's guide implies total budget contraction or just reallocation. That distinction determines whether this is cyclical or structural.
"The AI capex cycle poses multi-quarter distortion to software budgets, not a permanent drag; ROI realization could re-rate margins if productivity improves; monitor total IT budgets and CFO signals for evidence of direction."
Gemini overemphasizes a permanent OpEx-to-CapEx shift. The immediate risk is a multi-quarter capex-driven distortion: if AI infrastructure spend remains elevated, software investments may stay pressured until productivity proves ROI and SaaS churn or expansion responds. But if the productivity gains materialize, margins can re-rate. Don't ignore total IT budgets—growth or a quick rebound matters more than a 18-24 month deferral. My main point: monitor IBM's guidance and CFO commentary for evidence of total IT budget direction.
The panel agrees that IBM's Q2 warning reflects a temporary shift in capex towards infrastructure, with software budgets expected to recover once infrastructure bottlenecks ease. However, there's disagreement on whether this is a cyclical or structural issue, and whether it poses a sector-wide risk or a company-specific challenge for IBM.
Software recovery once infrastructure investments stabilize and AI productivity gains materialize, potentially leading to a rebound in software investments and margins.
Prolonged crowding out of enterprise software budgets by hyperscalers' capex growth, or a permanent shift from OpEx to CapEx spending, leading to margin pressure for SaaS firms.