AI Productivity Gains Fail to Translate Into Profit for Most Firms

Productivity up, profits flat

Artificial intelligence tools are making individual workers more productive, but most organizations have yet to capture meaningful financial returns from their AI investments, according to parallel research from McKinsey & Company and Deloitte.

McKinsey's 2026 State of AI survey found that 80% of respondents reported AI has improved their individual productivity, and half said it helps them make better decisions. Yet only 37% said AI has contributed to their organization's earnings before interest and taxes—a figure essentially unchanged from the previous year. The share of "AI high performers"—organizations attributing at least 5% of EBIT to AI—has remained flat at roughly 6% of respondents, according to the survey published in August 2026.

Deloitte's State of AI in the Enterprise 2026 report, based on responses from 3,235 leaders across 24 countries surveyed between August and September 2025, found a similar pattern. The firm reported that 37% of organizations are using AI at a surface level with minimal process changes, 30% are redesigning key processes around AI, and 34% are using AI to transform products, services, or business models.

Why it matters

The persistent gap between individual productivity gains and enterprise financial impact suggests that deploying AI tools without redesigning surrounding workflows, governance structures, and measurement systems may limit returns. For technology and business leaders evaluating further AI investment, this disconnect indicates that implementation strategy—not just tool selection—determines whether productivity improvements reach the bottom line.

Cost and procurement pressures

AI-related operating costs, including token expenses, have constrained AI use at about 20% of organizations surveyed by McKinsey, even as 60% expect to increase AI investment over the next year. The survey also found that 32% of respondents said their organizations had decided against purchasing at least one software product because it could be built internally using agentic coding tools—a pattern most common in technology and healthcare sectors.

On deployment scale, McKinsey found that 40% of respondents from organizations with more than $1 billion in annual revenue reported scaling AI agents in at least one function, up from 27% a year earlier. At smaller organizations, that figure remained flat at 22%.

Governance and talent gaps

Deloitte reported that while 42% of surveyed organizations view their AI strategy as highly prepared, preparedness ratings were lower for infrastructure, data management, risk and governance, and talent. Only about one in five organizations has a mature governance model for autonomous AI agents, according to the research.

On workforce response, Deloitte found that education aimed at raising general AI fluency—cited by 53% of respondents—was the most common talent strategy adjustment, ahead of measures such as redesigning roles or career paths.

The divergence between worker-level gains and enterprise outcomes, as documented in both surveys, points to a need for organizations to review how AI costs are tracked, how software renewal decisions account for internally built alternatives, and whether governance structures exist for AI agents already in use before expanding deployment further.

These findings were first reported by MarketScale, drawing on research from McKinsey & Company and Deloitte's AI Institute.