Fifty-seven percent of enterprises now have AI embedded in core business processes, up from 35% a year ago. Kyndryl’s second annual People Readiness Report, published July 17, surveyed 1,100 senior business and technology leaders across eight countries.
Only 32% of organizations with broad AI deployment reported achieving at least one of their top two AI objectives. Eleven percent had achieved both. Those results describe progress against selected objectives; they do not tell us that every other deployment produced no measurable value.
Only 23% of business leaders believed their people were fully prepared for AI, down six points from 2025. That decline in reported confidence accompanied broader deployment. The survey does not by itself establish that deployment caused the decline.
Read the workforce measures separately
The Achievers Workforce Institute’s 2026 State of Recognition Report offers related measures: 19% of workers felt confident using AI tools, and 18% felt supported in adapting to them. These are separate findings. Without the overlap between the groups, they cannot establish how many workers lacked both confidence and support.
Training needs to address the work people are expected to do with the tools. Role descriptions, handoffs, review duties, and escalation paths give that training a practical setting.
Gartner forecasts worldwide AI spending of $2.52 trillion in 2026, a 44% year-over-year increase. That spending forecast supplies context for the investment; it does not measure workforce readiness.
Examine the Pacesetter practices
Kyndryl identifies roughly 9% of respondents as Pacesetters. They are 1.5 times more likely to achieve AI-related revenue growth and 1.6 times more likely to report improved innovation.
The report highlights four practices among this group: redesigning roles around AI before deployment, structured change management alongside technical work, early governance guardrails, and ongoing workforce preparation.
The governance work includes decision-rights policies defining what AI can and cannot do, along with monitoring registries. Pacesetters are roughly twice as likely to have fully implemented AI governance as peers. Only one-third of organizations in the study have fully implemented training focused on working alongside AI.
These findings give a team practices to investigate. They leave open which changes would help a particular organization and how much of the reported outcome difference each practice explains.
Connect trust to a defined workflow
Eighty-one percent of organizations expect AI agents to make impactful business decisions within the next year. Only 25% currently have complete trust in autonomous AI systems, and only 27% use registries and monitoring across all AI systems.
Kyndryl reports higher employee confidence and better outcomes in organizations with stronger governance frameworks. For an operating team, that is a reason to inspect its own decision rights and visibility before expanding an agent’s authority.
Choose a deployed workflow and compare its current role descriptions with what people actually do. Identify who reviews the output, who resolves exceptions, and what employees need to learn. If the team cannot point to achieved objectives, check whether the objectives and measures are clear enough to evaluate, then test a specific change to the work.
Review the result with the people using the system. That can reveal a missing handoff or training need without assuming that every organization outside the Pacesetter group made the same mistake.