Enterprise AI Scaling Stalls as EBIT Gains Plateau in 2026

By Rebecca Stone

Only 22% of surveyed enterprises with annual revenue of at least $50 million had scaled artificial intelligence (AI) across multiple business units by 2026.

The shift from pilots to production affected enterprise workforces in unanticipated ways during 2026. Profitability gains from AI remained stagnant across sectors.

Everett Bishop, associate site editor for TechTarget's AI & Emerging Tech group, wrote: “AI has found its foothold within enterprises, but without specialized tooling, its progress will continue to plateau.”

Where does enterprise AI use stand?

The U.S. Census Bureau measured active AI use among U.S. firms. Its Business Trends and Outlook Survey (BTOS) covered December 2025 through May 2026.

• Large firms with at least 20 employees led active AI use among employees. Information, finance and insurance led U.S. enterprise AI use.
• The BTOS survey found that 20% to 23% of firms planned AI use by the end of 2026.
• McKinsey & Company surveyed 1,719 global respondents for “The State of AI in 2026: On the road to ROI.” The company found 89% used AI in at least one business function.
• Just over 50% of enterprises with at least $1 billion in revenue reported organization-wide AI scaling.
• Deloitte surveyed 3,235 global leaders for “State of AI in the Enterprise: The untapped edge.” Only 25% moved 40% or more AI experiments into production.
• Gartner surveyed 1,303 respondents at enterprises with annual revenue of at least $50 million. Only 22% successfully scaled AI across multiple business units.
• Most companies piloted or scaled AI in 2026. Scale became the standard deployment model.

This combination of broad usage, limited production conversion and concentrated scaling defines the 2026 bottleneck, not a simple failure to experiment widely with the technology.

Where is agentic AI heading?

Agentic AI (systems that perform tasks with limited human intervention) drew C-suite attention in 2026. Deployment plans grew faster than governance maturity.

• Deloitte found that 74% planned agentic AI deployment within two years.
• PwC polled 300 senior executives for its “AI Agent Survey.” Its leading agentic use cases closely matched Gartner's three most popular overall AI use cases:
  • Cybersecurity threat detection and response.
  • IT service desk automation.
  • Automated code generation.
• IBM's “2026 CEO Study: Rewiring the C-suite” found that 72% of CEOs predicted agentic AI would become a primary growth driver by 2030.
• Deloitte found that only 21% of companies had a mature governance model for autonomous agents.

The demand signal is broad, yet only 21% of organizations report mature governance for autonomous agents, exposing a material control gap as deployment plans advance.

Which risks constrain deployment?

Executives faced technology, strategy and sovereignty risks as AI moved into production. Deloitte identified five pressing C-suite risks.

• Gartner's third-quarter 2026 “Quarterly Emerging Risk Report” drew responses from 308 enterprise risk management leaders. It flagged two new AI-related risks. The leaders' concerns included:
  • AI discovery of cyber vulnerabilities.
  • AI strategy miscalculation.
  • AI-driven competitive displacement.
• McKinsey found that high-performing enterprises were more likely to mitigate AI-driven exploitation of vulnerabilities.
• Deloitte and IBM found that 83% of business leaders considered AI sovereignty (control over AI models, data and infrastructure) at least moderately important, if not essential.

Risk teams recognize both offensive discovery and defensive mitigation, yet governance evidence shows that organizations have not converted that awareness into mature oversight for autonomous agents.

Has AI produced measurable returns?

Return on investment (ROI) remained the sharper constraint. Adoption broadened faster than financial gains.

• Deloitte found that 74% of organizations hoped to grow revenue through AI. Only 20% achieved that goal.
• Google Cloud and the National Research Group surveyed 2,403 executives globally for “From token-maxxing to ROI: A reality check for enterprise AI.” Only 26% reported accelerating year-over-year returns. Google Cloud called them “AI ROI leaders.”
• McKinsey found that 37% attributed some impact on EBIT (earnings before interest and taxes) to AI. That share plateaued starting in 2025.
• IBM found that nearly 80% of executives expected AI to drive significant revenue by 2030. Only 24% knew where that revenue would come from.
• Deloitte found that 84% of organizations increased AI investment. It also found that 78% of leaders reported greater confidence in AI.
• Nearly 20% of organizations reported inconsistent, plateauing or declining AI ROI. Most still attributed some return to AI.

Investment and confidence therefore rose faster than proven returns, leaving enterprise AI with broad strategic commitment but only narrow and uneven financial results across sectors and business units.

What changes in 2027?

Large enterprises led adoption, scaling and returns. The data showed no significant adoption gap with