IBM targets 40% productivity gain with new AI model

International Business Machines Corp. is shifting the enterprise AI debate from building bigger models to rewiring business operations, announcing a new suite of tools aimed at delivering a 40 percent productivity increase for clients by 2030. The strategy, unveiled at its Think conference in Boston, positions IBM not as a direct competitor to foundation model developers, but as the essential orchestration layer for enterprises to deploy and govern AI agents at scale.

"In the next year or two, the enterprise world will sort into two camps: companies where AI runs their business, and companies where AI is still a project," IBM CEO Arvind Krishna said. "The line between those companies that make broader use of AI and those that don’t won’t simply come down to technology. It will be their operating model."

To enable this shift, IBM announced a new version of watsonx Orchestrate, a control plane to manage and audit AI agents from multiple vendors, and IBM Bob, a development environment for creating new agents. The company showcased its own internal success, where a bot called "Ask HR" reduced the touchpoints for generating an employee verification letter from 18 to just one. This end-to-end process redesign is what IBM intends to sell to its clients.

The move comes as IBM seeks to translate AI adoption into financial results. While the company reported a first-quarter revenue of $15.92 billion, its stock has faced pressure amid broader software market concerns. IBM is betting that by focusing on the complex task of enterprise integration and workflow automation, it can carve out a durable, high-margin business distinct from the crowded model-building space.

From Light Bulbs to Electric Motors

Krishna compared the current state of enterprise AI to the early days of electricity, arguing most companies are still just using it for "light bulbs"—small, useful tasks like summarizing emails or preparing for meetings. "It’s useful, but it’s not really redefining how the company runs," he said. The goal of IBM's new AI operating model is to move to the "electric motor" phase, where AI powers the core "factory" of the business, fundamentally changing production and efficiency. This involves redesigning processes from the ground up to remove human touchpoints, which Krishna argues is the key to unlocking significant returns. "I don’t begin with eliminating steps," he said. "I begin with how many touch points can I take out?"

The Hard Part of Implementation

Elevance Health, a major insurance provider and IBM customer, exemplifies the challenges and opportunities. The company is using AI for claims processing and to power digital assistants that can answer complex questions about benefits. Chief Digital Information Officer Ratnakar Lavu said the successful deployment of such tools requires a deep, sustained collaboration between business and technology teams, along with rigorous, transparent governance. Lavu noted that while the company is seeing clear ROI from individual AI-powered processes, the "connectivity of processes to see the net outcome" is where the hard work remains. This highlights the precise challenge IBM's watsonx Orchestrate, designed to manage agents from different providers including those from competitors like OpenAI, aims to solve.

IBM's strategy is a long-term play that hinges on convincing large enterprises to undertake difficult and expensive internal restructuring. By providing the tools and a blueprint for this transformation, IBM seeks to become the central nervous system for the AI-powered enterprise. The company also announced the general availability of its Sovereign Core offering, allowing for on-premise, air-gapped AI environments, further catering to the security and governance needs of large institutions.

This article is for informational purposes only and does not constitute investment advice.