Agentic AI is here. Is your workforce ready?

At Think 2026, IBM CEO Arvind Krishna drew a clear line in the sand: “The enterprises pulling ahead are not deploying more AI; they’re redesigning how their business operates.”

For two years, generative AI has been about experimentation, including chatbots, summaries and pilots that have not scaled. Agentic AI changes the equation. Autonomous agents don’t just assist; they act. They orchestrate workflows. They make decisions. They hand off work across systems and teams.

This shift is not a future state. It is happening right now. And it demands a fundamental question that most organizations have not yet answered: The technology is ready. Is your workforce ready to work alongside it?

The IBM IBV 2025 CEO Study surveyed 2,000 CEOs across 33 countries. The study found that only 25% of AI initiatives have delivered expected ROI over the last few years. And just 16% have scaled AI across the enterprise.

The data is clear: 85% of CEOs expect AI efficiency investments to pay off by 2027. But expectation is not a strategy. Many leaders are investing heavily without fully understanding the value of the technology or its impact on their people.

Here’s the disconnect: AI is being treated as a technology issue when it is increasingly a workforce issue. Buying the platform is the easy part. Building the skills to orchestrate agents, govern AI systems and redesign work around them is far more difficult. 

The organizations that succeed will be the ones that treat workforce readiness as a strategic capability, with business leaders and HR working together to build the skills the AI era demands.

If workforce readiness is the barrier, what does readiness look like in practice? When AI began handling entry-level work, many assumed that the answer was to cut early-career hiring. IBM made the opposite choice.

Nickle LaMoreaux, IBM’s Chief Human Resources Officer, recently announced that IBM is tripling its entry-level hires for the very roles some predicted AI would eliminate.

But these jobs are not the same jobs that existed three years ago. LaMoreaux has been clear: two to three years ago, AI could do most entry-level work. So IBM rewrote the job descriptions entirely.

A junior software developer now spends less time on routine coding (which AI handles comfortably) and more time with clients. An entry-level HR staff member steps in when chatbots fall short, correcting AI outputs and escalating to managers instead of fielding every employee query themselves. This is what workforce readiness looks like at scale: redefining roles, not eliminating them. 

LaMoreaux warns of a looming risk that every CIO should hear: cutting early-career hiring might save money in the short term, but it creates a shortage of mid-career talent down the line. That forces companies to poach experienced talent from competitors, which is a more expensive route with longer onboarding and weaker cultural fit.

Arvind Krishna has been clear on the underlying philosophy: “When people become more productive, organizations don’t shrink; they grow.“ At IBM, automating routine roles did not lead to contraction. It created capacity to hire more developers, strategists and client-facing talent.

So, what does workforce readiness mean specifically for the IT teams that will build, deploy and govern these agentic systems? Several specific capabilities are emerging as essential for IT readiness:

The focus has shifted from training models to deploying multi-agent LLM systems that work together. IT teams now need proficiency in LangChain, LlamaIndex, and AI-powered IDEs like Cursor and Windsurf. The role of the traditional developer is evolving into the AI engineer—someone who orchestrates and validates the outcomes of developer and tester agents, rather than writing every line of code themselves.

As agents move from batch to real-time, infrastructure skills are shifting from traditional MLOps to LLM operations (LLMOs), vector databases and edge AI deployment. Your IT organization needs engineers who understand how to serve models at the edge, not just in the cloud.

A new role is emerging called the context engineer. A context engineer is someone who designs and optimizes prompts to get accurate, reliable outputs from AI models and translates business needs into clear instructions AI can execute. It also tests and refines prompt techniques across teams. This role is not a nice-to-have. It is a core competency.

At level 4 of the agentic SDLC framework, governance must be embedded as policy-as-code, not bolted on after the fact. Your IT organization needs skills in jailbreak defense, constitutional AI and red teaming. These skills are not traditional security skills applied to AI. They are AI-native security skills.

Several roles now exist that did not exist two years ago: the agent governance lead (defining policies for responsible AI use) and the head of AI platform (designing the architecture and integrating AI with enterprise systems). The agent product owner is also a new role, owning the end-to-end user experience and ROI of agent-driven features. 

These roles require a combination of technical depth, business acumen and risk management that most IT organizations do not currently have.

The data confirms the urgency. According to the IBM IBV study, 54% of CEOs are now hiring for AI-related roles that did not exist a year ago. Entire job categories are being invented in real time. Nearly one-third of the workforce will need to reskill in the next three years to remain competitive.

This is why IBM is tripling entry-level hires and rewriting job descriptions. Prompt engineering, agent orchestration, policy-as-code and AI safety are new skills that cannot be hired off the street. They must be built. The organizations that will win are the ones that build a skills development machine, not just a hiring machine. 

Workforce readiness is no longer an HR initiative on the side; it is a business transformation priority. Here is where IBM can help.

IBM brings together the technology, consulting expertise and workforce transformation capabilities needed to operationalize agentic AI responsibly and at scale. From AI platform architecture and agent orchestration to skills development, governance frameworks and organizational redesign, IBM helps enterprises move beyond experimentation into measurable business outcomes.

Because the challenge ahead is not simply deploying more AI—it is building an organization ready to work with it.