IBM’s Strategic Shift in AI and Quantum Computing
Arvind Krishna, the CEO of IBM, recently discussed the company’s strategic direction in the fields of artificial intelligence and quantum computing. In an interview, Krishna articulated IBM’s focus on specialized AI models tailored for specific business applications, and the company’s long-term commitment to making advancements in quantum computing.
IBM, once a titan of the computing world, played a crucial role in the development of modern personal computers and even created one of the first AI systems to defeat a human chess champion. However, the company has shifted its focus. Instead of developing large consumer-facing products, IBM now concentrates on business-to-business solutions.
“We are a B2B company, and explaining what we do to the average reader—we’ll take all the help we can get,” Krishna quipped.
AI Strategy: Specialization Over Generalization
While IBM doesn’t develop massive, generalized AI models like those of OpenAI or Google, it does build smaller AI models designed for specialized, high-accuracy applications. This approach raises a critical question: will the economic gains from AI mostly accrue to companies that train massive “foundation models”? Or will they flow to companies, like IBM, that develop models for specific, accurate use-cases? Krishna suggests that is a false dichotomy.
Krishna explains that the company chose to focus on smaller, more specialized models because they can be purpose-built. In situations that demand extremely high accuracy, a model must be designed specifically for that situation.
“If you’re willing to have an answer that’s only 90% accurate, maybe. But if I’d like to control a blast furnace, it needs to be correct 100% of the time,” Krishna stated.
Quantum Computing: An Engineering Challenge
IBM has been investing in quantum computing for over a decade. Krishna sees quantum computing not merely as a scientific problem but an engineering one.
“We came to the conclusion that it’s an engineering problem more than it’s a science problem,” he explained, referencing the current issues with quantum computing
These key issues include high error rates and the brief periods during which quantum systems can maintain “coherence.” IBM believes that the field will advance rapidly over the next few years. If IBM can achieve some major technology milestones, the company will see an unparalleled market position.
Krishna believes that quantum computing will be additive, evolving in a similar way to previous technological changes, and he suggests IBM may occupy a central position like it did with mainframes and PCs.
