IBM CEO Arvind Krishna spoke to CNBC about the first-quarter earnings beat and factors facing the business.

The CEO mentioned the organization is maintaining guidance because of the Iran war and other uncertainty.

Krishna stated Anthropic’s Mythos AI model is powerful, but he expects others to catch up and remarked someone may have “already done it.”

International Business Machines CEO Arvind Krishna told CNBC on Wednesday that the Iran war and other geopolitical uncertainty are leading the firm to guide cautiously.

IBM beat analyst first-quarter earnings estimates on the top and bottom lines, but maintained guidance due to the macro uncertainty.

“Is there going to be an issue around oil as inflation goes up? Will that drive the public to spend a bit less? If they spend a bit less, it’s not a direct impact on me, but a lot of consumer companies are my clients, like Walmart. If individuals are buying less at Walmart, they’re going to find a way to control their costs, so then they’ll invest in less,” he mentioned.

Krishna noted that despite the Iran conflict, IBM’s Middle East business did well.

The corporation reported first-quarter revenue of $15.92 billion, beating the $15.62 billion consensus estimate from LSEG. Earnings per share came in at an adjusted $1.91, 10 cents better than expectations. Software beat, with Red Hat growth rebounding to 10%.

Krishna mentioned he is also cautious about growth concerns in Europe. This also touches on aspects of earnings report.

“That’s the only place where I think there is some squinting because it is also a little bit of jadedness, right? You had the Covid shocks, you had the Ukraine war. So they’ve gone through these shocks a few times, and they actually come out okay. I think this time around is an open question,” Krishna commented.

“I don’t think anyone will know the answer for another month or two,” he added.

Mythos Furthermore, experts in bull market note the continued relevance.

Anthropic’s launch two weeks ago of its powerful latest Mythos artificial intelligence model that is capable of finding security vulnerabilities at unprecedented speed and volume sent shockwaves throughout tech, but Krishna mentioned others will soon follow.

“Somebody does a thing. It looks magical. It looks wonderful. We think it’s the only thing. Three months later, somebody copies it and actually does it better,” he mentioned.

“I’d be surprised if somebody else hasn’t already done it but hasn’t bothered to claim it,” he added.

IBM shares plunged in April after the AI startup remarked its Claude Code tool could modernize legacy systems that run Common Business-Oriented Language. COBOL is a code system developed in the late 1950s that is regularly used in business data processing.

The release of Mythos triggered a surprise meeting between Treasury Secretary Scott Bessent and Fed Chair Jerome Powell with the heads of the top U.S. banks over AI cyber concerns. Bessent and Vice President JD Vance held a call with tech CEOs like Anthropic’s Dario Amodei, OpenAI’s Sam Altman, xAI’s Elon Musk and others about the same matter.

“It’s a very huge conversation, and there is no question that it can find and exploit vulnerabilities at a rate and pace that has not been seen so far,” Krishna noted.