IBM’s AI business hits $9.5B as mainframe sales jump 17%

IBM beat Wall Street expectations for third-quarter revenue and profit, driven by surging demand for its software and new mainframe systems as businesses rush to upgrade their infrastructure for AI deployment. The company posted revenue of $16.33 billion and adjusted earnings of $2.65 per share, both exceeding analyst estimates, though shares fell nearly 4% in after-hours trading despite the strong results.

What you should know: IBM’s software segment delivered exceptional growth, rising 10% to $7.21 billion and beating estimates as enterprises prioritize platform upgrades for AI capabilities.

• The infrastructure segment, which houses IBM’s mainframe business, saw revenue surge 17% to $3.56 billion during the quarter.
• IBM’s AI book of business grew to $9.5 billion, up $2 billion from the previous quarter, with consulting revenue comprising about $7.5 billion of this figure.
• The company raised its full-year revenue outlook, now expecting growth of more than 5% at constant currency, up from its prior forecast of at least 5% growth.

The big picture: Businesses are investing heavily in upgrading their systems to handle data-intensive AI applications, creating a boom for IBM’s high-margin software and infrastructure products.

• The demand reflects companies’ strategic shift toward AI-ready platforms, with IBM positioned to benefit from this multi-year technology transition.
• IBM’s stock has risen about 30% this year, though investors appeared to take profits following the earnings announcement.

Key details: IBM’s latest mainframe iteration is powered by chips specialized for AI applications and has seen strong adoption across industries.

• The financial sector is driving significant demand for the new mainframe, which allows banks to maintain strict data residency and encryption requirements while adopting AI technology.
• The mainframe follows a three-year upgrade cycle, with the previous version having completed its run.

In plain English: A mainframe is essentially a high-performance computer designed to handle massive amounts of data and complex calculations simultaneously. Think of it as the industrial-strength engine that powers large organizations’ most critical operations, from processing millions of bank transactions to managing airline reservations worldwide.

What they’re saying: CFO Jim Kavanaugh emphasized how client needs are driving purchasing decisions across IBM’s portfolio.

• “Addressing client pain points, around AI, around sovereignty … that is driving the client buying behavior overall,” Kavanaugh told Reuters.
• The focus on data sovereignty reflects growing regulatory and security concerns as companies implement AI systems.

Competitive landscape: IBM’s consulting revenue grew about 3% to $5.32 billion as clients prioritized spending on large-scale AI-linked consulting projects.

• The company’s AI book of business metric combines both bookings and actual sales across various products, providing a comprehensive view of AI-related demand.
• However, the AI book of business figure notably excludes mainframe sales, suggesting IBM’s AI revenue could be even higher when including all AI-capable infrastructure.