Investing.com -- International Business Machines (NYSE:IBM) CEO Arvind Krishna said this week that artificial intelligence has already replaced several hundred jobs in human resources at the company, even as AI deployment fuels headcount growth in other areas. “Our total employment has actually gone up,” Krishna told The Wall Street Journal, as AI investments free up resources for higher-value roles.
The comments, made while IBM hosts its Think 2025 conference in Boston, emphasized that internal AI adoption has allowed the company to redirect hiring efforts toward domains requiring critical thinking, such as software engineering, sales, and marketing. Krishna said the company’s focus remains on activities that “face up or against other humans, as opposed to just doing rote process work.”
In line with broader enterprise trends, IBM is prioritizing smaller, domain-specific generative AI models that offer cost and performance advantages over large-scale language systems. “Smaller models are incredibly accurate,” Krishna noted. “They’re much, much faster. They’re much more cost effective to run.”
Unlike massive foundation models like GPT-4, IBM’s Granite 3.0 family, ranging from 3 billion to 20 billion parameters, is optimized to be deployed and tuned in enterprise environments. These models are now embedded in WatsonX, IBM’s AI platform that includes tooling for building, training, and deploying AI at scale.
Krishna said only around 1% of enterprise data is currently being accessed by generative AI, due to fragmentation across cloud, edge, and data center systems. To close that gap, he emphasized that future AI solutions must be integrated, configurable, and tailored. “Success is going to be defined by integration and business outcomes,” Krishna said.
IBM’s WatsonX Orchestrate platform currently offers 150 pre-built models and allows customers to build their own agents in minutes. The platform is designed to work alongside the clients’ preferred AI stack, continuing IBM’s trend of vendor-agnostic solutions.
Internally, IBM’s use of AI agents mirrors the value proposition it now offers customers. Krishna shared that client interest remains strong across functions including HR, retail, and manufacturing, where smaller, specialized agents provide the clearest returns.
IBM also sees minimal exposure to global tariff shifts, citing its $150 billion cumulative U.S. investment and domestic manufacturing footprint. While AI’s labor impact is unfolding unevenly across industries, Krishna said IBM’s example shows how automation can increase staffing in strategic functions over time.