Arvind Krishna is sending a clear message on the next phase of the artificial intelligence race. The chief of International Business Machines (IBM) said that the bigger risk for companies right now isn’t moving too fast, it’s falling behind.
In an interview with Semafor posted on April 17, Krishna said many business leaders still don’t fully realize how important technology has become to their companies. That gap, he suggested, could shape who wins and who loses in the years ahead.
The comments come as IBM shifts toward higher-growth areas like AI and hybrid cloud (which connects users’ own systems and public cloud services). The company has been building out its watsonx AI platform and expanding enterprise AI services, while also investing heavily in longer-term bets like quantum computing.
Shares of IBM are down 14% year-to-date as of writing. The stock experienced a sharp decline of more than 20% in February, driven by investor fears that new AI technology from Anthropic (Claude Code) could automate legacy COBOL system modernization, threatening IBM’s high-margin consulting services and mainframe ecosystem.
IBM is set to post its first-quarter results on Wednesday (April 22).
Getty Images
Morgan Stanley analyst Erik Woodring lowered IBM’s price target to $215 from $247 with an equal weight rating before the earnings, The Fly reported.
The firm thinks Q1 numbers should move higher, but it doesn’t believe they are “enough to reignite the IBM bull case,” as software, services, and quantum peers continue to de-rate.
IBM CEO says many companies still underestimate AI’s impact
Krishna said the biggest issue he sees is that many executives still treat technology as a support function, rather than something central to how a business operates.
“I think that most CEOs have not yet woken up to [the fact that] technology is probably as important as their balance sheet,” he said.
Related: Cathie Wood buys $2.5 million of tumbling megacap stock
He described this as part of a longer shift in how companies create value. In the past, the focus was primarily on physical assets, then talent, then access to capital. Now, he said, technology is just as critical.
“How well you can use technology to run your company is the most important,” he added.
Krishna also pushed back on the idea that companies should be cautious about adopting AI too quickly. In his view, hesitation is the bigger danger.
“I never have a fear of moving too fast,” Krishna said.
“The fear is of falling behind or of making completely the wrong bet or the wrong strategy,” he added.
IBM bets on a mix of AI models and Quantum
On the technology side, Krishna said the future of AI won’t be dominated by a single system or platform.
“We are in a world where large language models are extremely intriguing and of a lot of value, but not for all things,” Krishna said.
Related: BNP Paribas delivers blunt message on Microsoft stock
“We are going to have three or four very large models. They do certain things,” he said. Alongside those, he expects many smaller, specialized models built for specific tasks.
“We’re going to have hundreds, maybe thousands of smaller but far more purpose-built models, with much more curated data,” he added.
That approach aligns with IBM’s broader strategy. The company has pushed a hybrid model in cloud and AI, where clients mix different systems instead of relying on one provider.
Looking further out, Krishna said quantum computing is the next big area to watch.
“Quantum is probably 3 to 5 years away in terms of timing,” Krishna said. “It’s going to solve the kinds of problems AI could not do.”
“AI is great at predicting a bit of the future…Quantum computes the future. So that unlocks problems you can’t even imagine doing with AI,” he said.
IBM has been working on quantum technology for years, and Krishna said companies should already be thinking about how it could apply to their business, even if large-scale adoption is still a few years away.
“They need to start thinking about it now,” he said.
Related: Key quantum computing stock jumps 20% in a day, here’s why
