Half of enterprises investing in AI see no meaningful returns. Kyndryl says the problem isn't the technology.
Nearly half of enterprises spending heavily on AI report struggling to see real results, according to Kyndryl's Readiness Report. An IBM study backs this up: just 25% of AI initiatives deliver expected return on investment, and only 16% have scaled across the organization.
The infrastructure company launched Agentic Service Management last month to address why AI pilots stall before reaching production scale. The framework combines a maturity model, structured assessments, and implementation blueprints designed to move organizations from traditional service operations to autonomous, intelligent workflows.
The core argument: most enterprise environments were built for people managing tickets and tools, not for fleets of autonomous agents executing tasks across hybrid and multi-cloud systems. That mismatch is what's stopping AI from moving beyond pilots to actual outcomes.
Where organizations actually stand
Kyndryl Consult starts by assessing where an organization sits across four areas: service management, AI governance, security, and operations. The assessment benchmarks current capabilities against emerging standards-including ISO 42001, the AI management standard most enterprises haven't yet engaged with.
From there, Kyndryl delivers a gap analysis and a phased roadmap. A separate offering, Kyndryl Agentic AI Digital Trust, provides a security-first framework for governing agentic AI deployments across hybrid and multi-cloud environments.
For regulated industries-financial services, healthcare, public sector-this governance piece is urgent. An AI agent operating outside defined boundaries can quickly become a compliance problem.
Rather than leading with technology, Kyndryl is leading with organizational readiness. Many vendors sell AI agents. Fewer address what the business itself needs to look like before those agents can operate reliably at scale.
Testing it internally
Kyndryl is applying Agentic Service Management to its own internal service delivery operations through Kyndryl Bridge, an open integration platform. The company already runs nearly 200 million automations per month across more than 8,000 certified playbooks.
That existing foundation means the maturity model is grounded in running mission-critical infrastructure at enterprise scale, not assembled in a strategy document. For customers doing due diligence, that distinction matters.
The operating model is the bottleneck
The market is shifting from AI strategy to AI execution. Two years of heavy investment left many organizations with a portfolio of pilots and flat ROI lines.
Kyndryl's position is that governance, workflows, and controls are still rooted in the pre-AI era. The operating model, not the technology, is the real constraint.
For a managed services company, this positioning shifts the conversation from platform selection to operational transformation-which is Kyndryl's core business.
The same governance principles apply to customer-facing work. As AI agents take on case resolution, proactive outreach, and self-service escalation, organizations need the same oversight. Companies building those internal controls now will be better positioned when customer-facing deployments scale and regulators start asking questions.
The real test
Maturity models have a long history of becoming shelf documents. Kyndryl has correctly identified the problem and built a structured way to address it.
Whether that translates into measurable outcomes for customers-or stops at a gap analysis and roadmap-is what enterprises should push hard on before signing up.
For management teams tasked with delivering AI outcomes, understanding your organization's readiness across governance and operations is a prerequisite. Learn more about AI Agents & Automation and how to approach AI for Executives & Strategy to build a foundation for scaling AI beyond pilots.
Your membership also unlocks:
