IBM’s CEO, Arvind Krishna, asserts that global trade is alive and well despite the challenges posed by the Trump Administration’s criticisms of globalization. He believes that the U.S. must foster international trade to drive economic growth.
“I am a strong advocate for global trade—a concept that dates back to economists of the 1800s,” Krishna emphasized during an interview at SXSW on Tuesday. “The idea is that a 10% rise in global trade correlates with a 1% boost in local GDP. If we truly want to maximize local growth, global trade is essential.”
Krishna noted that global trade is linked to welcoming international talent to the U.S. He criticized the Administration and its supporters for proposing stricter regulations on student and H-1B work visas, which they argue could disadvantage American workers.
“We should encourage talented individuals to come here and use their skills,” Krishna stated. “While it’s vital to cultivate our own talent, we can’t do that effectively without learning from the world’s best. The U.S. should aim to be a hub of international talent, supported by relevant policies.”
During the extensive discussion, Krishna also delved into topics like geopolitics and artificial intelligence, which he views as an essential technology—but not a cure-all. He disagreed with a recent assertion from Dario Amodei, the CEO of Anthropic, who predicted that AI would be responsible for 90% of code in the coming months.
“I believe the actual figure will be closer to 20-30% of code generated by AI—not 90%,” Krishna said. “There are simple applications where that may be true, but there are equally complex cases where AI won’t contribute at all.”
Krishna expressed confidence that AI will enhance the productivity of programmers, leading to increased outputs for both employees and employers, countering fears that AI could eliminate programming positions.
“If you can produce 30% more code with the same workforce, will that result in more or less overall code?” he posed. “History indicates that the most productive companies capture more market share, allowing for greater product output and further market expansion.”
However, it’s important to note that IBM has a particular stake in portraying AI as a non-threatening technology; the company offers various AI-driven products and services, including coding assistance tools.
This perspective marks a shift for Krishna, who stated in 2023 that IBM would pause hiring for roles that AI could potentially replace.
Krishna likened the discourse surrounding AI’s impact on employment to earlier discussions about calculators and Photoshop affecting mathematicians and artists. He acknowledged ongoing challenges related to intellectual property in the context of AI training and its outputs, yet ultimately views the technology as a beneficial and enhancing force.
“AI is a tool,” Krishna remarked. “If it enables everyone to produce higher quality work, consumers benefit from better-quality products as well.”
Krishna anticipates that the costs associated with AI will decrease. While he recognized that current models, like OpenAI’s, require significant computing power and energy, he believes future advancements will enable AI to utilize “less than 1%” of today’s energy consumption through innovative techniques like those demonstrated by the Chinese AI startup DeepSeek.
“DeepSeek shows us that it’s possible to operate with much smaller models,” Krishna noted. “However, the lingering question remains: do we still need large foundational models? That aspect seems to be what [DeepSeek] did not address.”
Despite AI’s predicted commoditization, Krishna remains skeptical about its potential to unlock new knowledge for humanity. He echoes sentiments expressed in a recent essay by Hugging Face co-founder Thomas Wolf, positing that quantum computing—a field in which IBM is heavily invested—will be pivotal in accelerating scientific advancements.
“AI learns from existing knowledge, literature, graphics, etc.,” Krishna asserted. “It does not seek to discover new knowledge. I am not convinced that the current generation of AI will lead us to what is termed artificial general intelligence—a point where AI can hold comprehensive, reliable knowledge exceeding that of past luminaries like Einstein or Oppenheimer.”
Krishna’s views contrast sharply with those of OpenAI CEO Sam Altman, who, in an essay released earlier this year, suggested that achieving “superintelligent” AI could be within reach in the next few years, potentially leading to a significant acceleration in innovation.
Compiled by Techarena.au.
Fanpage: TechArena.au
Watch more about AI – Artificial Intelligence
