IBM Proves Old Tech Can Win At New AI

IBM (NYSE: IBM) beat earnings and proved its AI strategy is actually working. Q4 revenue hit $19.69 billion versus $19.21 billion expected—a 12.1% year-over-year beat. EPS came in at $4.52 versus $4.29 expected, a 5.4% beat. IBM’s Software business is leading this growth: +14% growth to $9 billion in Q4, driven by Red Hat, automation, and WatsonX AI platform. Infrastructure revenue surged 21% to $5.1 billion, with IBM Z mainframe sales up 67% year-over-year. The mainframe cycle is back, which is both impressive and slightly hilarious in 2026. See, IBM is pivoting to AI, and it is succeeding at that. IBM’s generative AI book of business hit $12.5 billion, more than doubling from last year. About 80% of that is consulting services, meaning enterprises are paying IBM to build production AI systems, not just run pilots.

That being said, if you seek an upside with less volatility than holding an individual stock, consider the High Quality Portfolio. It has comfortably outperformed its benchmark—a combination of the S&P 500, Russell, and S&P MidCap indexes—and has achieved returns exceeding 105% since its inception. Why is that? As a group, HQ Portfolio stocks provided better returns with less risk versus the benchmark index; less of a roller-coaster ride, as evident in HQ Portfolio performance metrics.

Image by Curtis Parfitt-Ford from Pixabay

How is that different from Meta and Microsoft burning billions?

IBM is capturing the services margin. While Meta and Microsoft spend $100+ billion building infrastructure, IBM charges companies to actually implement AI solutions. It’s a lower-risk, higher-margin business model. No speculation—just billable hours.

What about profitability?

IBM’s pre-tax margin (Non-GAAP) was 24.1%, down marginally from 24.3% last year but still healthy. Free cash flow hit $14.7 billion in 2025. For 2026, IBM expects free cash flow to grow by $1 billion to roughly $15.7 billion. That’s real cash generation.

What’s the guidance?

Revenue growth of “more than 5%” for 2026, decelerating from 8% in 2025. That’s below the double-digit growth Meta and Microsoft are posting, but IBM is being conservative. Analysts expected 4.6% growth, so this is slightly above consensus.

Also, IBM acquired Confluent for $11 billion in December 2025 to solve the “data-in-motion” problem—feeding real-time streaming data into AI models. This acquisition positions IBM for the shift from “chatbot AI” to “agentic AI” (autonomous AI agents that execute workflows).

Why did the market reward this?

Because IBM is showing proof of AI monetization. The $12.5 billion AI book of business is real consulting revenue, not deferred commitments. The mainframe cycle is boosting infrastructure. Free cash flow is growing. And the Confluent acquisition is strategic, not empire-building.

Valuation check—is $315 expensive?

At $4.52 quarterly EPS (TTM: $11.56), the P/E is about 27.5x. That’s not cheap for 5%+ revenue growth and $15 billion in annual free cash flow. IBM is trading at a lofty valuation for this growth profile. The current price now aligns with the $316 average of analysts’ price estimates. We currently have a $294 price estimate for IBM. While we will soon revise this to reflect the latest earnings, we don’t foresee significant upside from the current estimate. Why? Consider a stock like MSFT, which is seeing 37% growth for Azure and 17% company-wide, and 45% operating margins while trading at 30x trailing earnings. You can get all excited about IBM, but make no mistake: it is expensive. If you are willing to pay 27–30x earnings, there are far better options available in the market. See, investors are punishing MSFT for sky-high CapEx spending on data centers and chips, while IBM has a much smaller CapEx burden. While IBM is a “safer” capital-efficient play, paying nearly the same multiple for IBM’s consulting growth as you do for Microsoft’s infrastructure dominance is a tough sell.

What are the risks?

• Integration friction from the Confluent acquisition,
• Potential macroeconomic headwinds hitting enterprise IT spending,
• Competition from Microsoft Copilot, Amazon Bedrock, and Google Gemini in the enterprise AI space, and
• The upcoming IBM z17 mainframe cycle needs to deliver, or infrastructure revenue collapses.

So, IBM is the AI winner?

In enterprise production AI, yes. IBM isn’t trying to build the next ChatGPT. They’re building the plumbing for banks, manufacturers, and governments to deploy AI safely and compliantly. It’s a smaller, slower, but more profitable game than the hyperscaler AI arms race.

Final take?

IBM proved AI can be monetized through consulting and infrastructure rather than just burning capex. The 7% after-hours rally reflects relief that “old tech” can compete in the AI era. But the $315 price tag takes into account a lot of optimism and cash flow growth.

See, investing in a single stock without comprehensive analysis can be risky. Consider the Trefis Reinforced Value (RV) Portfolio, which has outperformed its all-cap stocks benchmark (combination of the S&P 500, S&P mid-cap, and Russell 2000 benchmark indices) to produce strong returns for investors. Why is that? The quarterly rebalanced mix of large-, mid-, and small-cap RV Portfolio stocks provided a responsive way to make the most of upbeat market conditions while limiting losses when markets head south, as detailed in RV Portfolio performance metrics.

Invest with Trefis Market-Beating Portfolios
See all Trefis Price Estimates