IBM CEO Arvind Krishna said global trade has not died despite the Trump administration’s attack on globalism. In fact, he believes that the key to US growth is accepting international exchanges.
“So, I’m actually a solid follower. I think you’ll go back to the economists who learned world trade in the 1800s. And I think their perspective will lead to a 1% increase in local GDP with every 10% increase in global trade.” “So, if you really want to optimize for your local (growth), you need to do global trade.”
Global Trade is working with allowing foreign talent to flow into the US, Krishna said. The administration and its allies are seeking increased student and H-1B job visa restrictions, and they argue that we are putting us at a disadvantage.
“We want people to come here, bring their talents to them and apply those talents,” Krishna said. “And we also want to develop our talent, but if we don’t bring the best people from all over the world to help our people learn, then you can’t develop it either. So we should be an international talent hub and we need a policy that aligns with it.”
In a wide range of interviews, Krishna touched on not only geopolitics but also AI. We consider this to be a valuable technique, but there is no panacea.
He opposed recent predictions from humanity CEO Dario Amodei. 90% of the code may be written by AI in the next 3-6 months.
“I think this number will lead to a possibility that 20-30% of the code will be written by AI, not 90%,” Krishna said. “Are there really simple use cases? Yes, but equally complex numbers will be zero.”
Krishna said he believes that AI will make programmers more productive and not eliminate programming jobs, but will increase the output of employers and ultimately make programmers more productive.
“If you can run 30% more code with the same number of people, do you write more code or get it less?” he said. “History shows that the most productive companies will gain market share, allowing them to gain more market share because they can produce more products.”
Certainly, IBM has a vested interest in presenting AI as a non-threat. The company sells a range of AI-powered products and services, including support coding tools.
The statement is also a bit of a reversal for Krishna, who said in 2023 that IBM suspended employment for back-office functions and said the company could replace AI technology.
Krishna compared the arguments surrounding AI with early discussions surrounding calculators and Photoshop, which replace mathematicians and artists. He acknowledged that there are “unresolved” challenges in intellectual property with regard to AI training and output, but ultimately, technology is a positive and augmented force.
“It’s a tool,” Krishna said of AI. “If using these tools and everyone produces better quality, then consumers are consuming better quality (products).”
Krishna predicted that the tool would be cheaper. He said that inference models like Openai’s O1 require a lot of computing and are therefore energy-intensive, but AI believes it uses “less than 1%” of the energy it uses today, thanks to new technologies like the ones demonstrated by Chinese AI startup Deepseek.
“I think Deepseek gave us a preview that we can live in a much smaller model,” Krishna said. “It still raises questions, but do you really need a big model to get started? And I think that’s something I didn’t talk about (deepseek).”
However, AI becomes commoditized, but Krishna is not sure it will help humanity arrive at new knowledge. Rather, Krishna believes quantum computing – technology IBM is investing heavily for nothing – is key to accelerating scientific discovery.
“AI learns from already produced knowledge, literature, graphics and more,” Krishna said. “It’s not trying to understand what’s coming… I’m someone who doesn’t believe that the current generation of AI will lead us to what is called artificial general information… When AI can have all the knowledge, we can answer questions beyond what Einstein or Oppenheimer, or all Nobel Prize ads are summed up.”
Krishna’s argument is in contrast to the people of Openai CEO Sam Altman who claims that “super intelligent” AI is within the realm of possibilities within the next few years and can dramatically accelerate innovation.