IBM may be the premier way to play the burgeoning quantum-computing market, but mounting threats to its consulting business have Wall Street feeling wary.

Susquehanna analyst James Friedman on Friday initiated coverage on shares of legacy tech giant International Business Machines at Neutral with a $303 price target. IBM stock was slightly higher at $295.68 in Friday's premarket session; Friedman's target suggests it could rise 2.5% from current levels.

"There are so many reasons to like IBM, which is why many investors do," Friedman wrote. Critically, his optimism hinges largely on IBM's quantum-computing business, which he calls "a great way to participate."

In a rare move, the analyst has pinned a dollar amount to IBM's quantum division, which sells cloud access to its vast fleet of quantum computers today. Friedman values it at $65 a share.

He believes IBM is positioned to capture a disproportionate share of a $650 billion quantum-computing market by 2040, adding that Big Blue "has the leading presence and has consistently delivered ahead of schedule."

Although quantum is increasingly important part of the business, investors can't overlook IBM's legacy consulting business. This, for Friedman, is where problems arise. IBM may be "a cheap quantum call, but consulting worries us," he wrote.

Indeed, well-documented anxieties over artificial-intelligence disruption of the software and consulting sectors have sent the market into a tailspin on multiple occasions this year.

While those concerns have failed to stick for some software stocks, IBM has been feeling the pressure. The stock had its best week in more than 20 years at the end of May, fueled by enthusiasm over its quantum prospects, but that rally wasn't enough to completely erase losses it suffered this year. Heading into Friday's session, IBM has fallen less than a percentage point in 2026. That puts it behind the benchmark S&P 500, which has gained more than 10% this year.

A massive 25% rally in the second quarter has brought IBM shares "within a whisper" of record highs, according to Friedman. Regardless, he sees problems ahead. The bulk of IBM's artificial-intelligence book of business sit in low-margin consulting, which he calls "a fragile end market where revenue dilutes consolidated margins."

Moreover, he believes the business faces pressure from data-center operators and AI-native companies. As hyperscalers bundle frontier AI models into existing cloud contracts at massive scale, this may relegate IBM's watsonx portfolio of AI products to niche workloads, capping its growth.

And then there's the COBOL problem. COBOL, a shorthand for Common Business-Oriented Language, is one of the oldest, high-level computer programming languages designed specifically for enterprises. Anthropic earlier this year released a tool to streamline the modernization of legacy COBOL codebases -- a release widely viewed as a harbinger of even more powerful tools to come.

As Friedman sees it, AI-assisted COBOL modernization -- the process of updating legacy mainframe applications to function in modern, cloud-native environments -- could make migration cheap, compressing IBM's high-margin mainframe annuity and consulting pricing power.

In short, "we see downside catalysts the consulting industry finds hard to shake," Friedman wrote. While short of a doomsday scenario, these challenges are enough to convince him to stay sidelined for now.

Write to Mackenzie Tatananni at mackenzie.tatananni@barrons.com

This content was created by Barron's, which is operated by Dow Jones & Co. Barron's is published independently from Dow Jones Newswires and The Wall Street Journal.

(END) Dow Jones Newswires

July 10, 2026 08:36 ET (12:36 GMT)

Copyright (c) 2026 Dow Jones & Company, Inc.