By Asa Fitch

As companies weave artificial intelligence into the fabric of their businesses, International Business Machines is making a case that it is relevant to investors again. But it still has a lot to prove.

Without much fanfare, Chief Executive Arvind Krishna has built up $6 billion of bookings around generative AI. This mostly consists of consulting contracts for companies trying to harness the technology. IBM's software business is also growing, and it has become more focused since the spinoff of its IT-outsourcing operation, now called Kyndryl, in 2021.

It all adds up to a company that is doing better than it has in a while, if not quite going gangbusters. Former IBM CEO Lou Gerstner wrote a book in 2002 about turning around the American business-computing icon in the 1990s: "Who Says Elephants Can't Dance?" Now, there is an argument that Krishna has IBM as close to dancing as anyone since.

Yet after years as a perennial laggard, now the stock has already priced in a growth transformation. That has set IBM up for disappointments like the one that struck Thursday, when the stock fell 6.6% despite reporting decent earnings.

Even after that sharp decline, IBM still trades at around 21 times forward earnings, compared with a 10-year average of just 12 times. That is actually a premium to where AI heavyweights such as Nvidia, Google parent Alphabet and Microsoft are trading, something IBM hasn't quite earned yet.

Since becoming CEO in 2020, Krishna has striven to make IBM into more of a facilitator of companies' tech transformations, dialing back on a practice of steering corporate customers only toward IBM products and services.

Central to the shift has been the company's $34 billion acquisition of open-source software giant Red Hat, a deal Krishna was deeply involved in before he took the reins.

Krishna correctly understood that many companies wanted an information-technology setup that included some of their own hardware and some that they rented out from large cloud-computing companies like Microsoft and Amazon. IBM positioned itself in the middle with Red Hat, developing and selling software to help companies manage such IT setups. Red Hat's revenue has about doubled since IBM bought it.

That makes IBM sound more like a growth company than it has in years. But IBM's results this week raised some concern about the health of its software business, which J.P. Morgan analysts said fell short of expectations in the first quarter when stripping out the impact of shifting currency valuations. And there is an even bigger cloud hanging over the cyclical consulting business, which fell 2% in the quarter.

Much of IBM's consulting work, which accounted for nearly a third of the company's $62.8 billion of revenue last year, has little to do with AI. And ironically for IBM, the prospect of AI shaking up how businesses work has led companies to put off big IT projects. That means they need less consulting help overall.

Layer on top of that a more uncertain economic outlook that could damp corporate spending, and it is hard to get overly excited about the consulting business, which Morningstar analysts expect to grow at an annual pace of about 4% in the coming years.

There's a broader problem with corporate AI, too. Companies are still largely in the experimentation phase, and it is unclear whether they will need as much of IBM's hand-holding as the technology matures. S&P Global Market Intelligence analyst William Fellows likened today's AI to an inefficient and fuel-hungry engine. Companies are spend-happy now -- venture-capital firm Menlo Ventures estimated it spent $13.8 billion on AI last year -- but a more efficient future could negate many of today's opportunities.

Worryingly, IBM doesn't have much of a record of capitalizing on big technology shifts. It didn't make significant inroads in cloud computing -- a corporate IT shift that should have been in its wheelhouse -- as Amazon's AWS and Microsoft's Azure came to dominate that industry. It also failed to convert past AI successes like the "Jeopardy!"-playing Watson, or chess-playing Deep Blue, into thriving businesses. Its Watson Health AI initiative fell short.

What's more, IBM lacks offerings such as Microsoft's corporate-productivity software or cloud-computing operations that it could use to deepen customer relationships built initially on AI consulting. That limits its scope to extract more revenue from the AI opportunity.

Krishna has certainly put some pep back into the step of a company many analysts and investors had written off only a few years ago. But it would be easier to shrug off the cyclical pressures if IBM were dancing toward AI prosperity in a straighter line.

Write to Asa Fitch at asa.fitch@wsj.com

(END) Dow Jones Newswires

April 25, 2025 05:30 ET (09:30 GMT)

Copyright (c) 2025 Dow Jones & Company, Inc.