We are in a world where AI is “supposedly” changing everything. But up until now, it’s been a fancy autocomplete or chatbot that sometimes loses the plot.
Fast forward: we’re moving into the era of agentic AI, and if you’re a program manager (PM) or a leader, the vibe of your daily job is about to shift. We’re not just talking about tools that help you write emails; we’re talking about digital teammates that actually do things.
What is “Agentic” AI anyway?
Agentic AI refers to systems designed to operate with a greater degree of autonomy than traditional AI tools. Traditional AI systems—such as large language models like ChatGPT or Gemini—typically respond to explicit user instructions by generating outputs based on the information provided. These systems generally require continuous user input to determine subsequent actions and often operate with limited contextual awareness.
Agentic AI is goal-oriented. Rather than responding to isolated prompts, it can be assigned a broader objective—such as reducing a project backlog while adhering to budget constraints—and independently determine the sequence of actions required to achieve that objective. You can give it a goal such as, “Hey, we need to clear the Q3 backlog and stay under budget,” and it goes off, looks at the data, talks to other systems, and actually executes the steps to make it happen.
Tools such as Claude demonstrate early implementations of this approach by allowing users to provide structured context, including files, datasets, and decision constraints. This seemingly small change—providing richer context—dramatically improves both the number of actions performed and the quality of the outputs.
It’s not surprising, then, that Gartner predicts that by 2028, about 15% of daily work decisions will be made by these autonomous agents[1].
Should Program Managers be excited or worried?
If you’ve ever spent your entire Monday chasing people for status updates or manually moving Jira tickets around, you know the “PM tax.” It’s the administrative grunt work that keeps you from actually strategizing. The shift underway is moving from task manager to decision integrator.
Imagine an AI agent that can flag a resource conflict, re-allocate the budget, and notify the stakeholders before you even finish your morning coffee. Your job shifts from clicking buttons to making sure the AI is moving in a direction that actually aligns with the company’s “big picture” goals. It can simplify the inputs from across different systems by consolidating the project timeline with all these updates.
This naturally raises the question: will the PM role become redundant with the rise of agentic AI? The answer is no. A program manager’s responsibilities extend far beyond Jira tickets and status updates. The role also includes strategic oversight of upcoming initiatives, in-depth analysis, and long-term alignment with key stakeholders.
As the landscape continues to evolve, the program manager’s focus shifts from administrative execution to strategically leveraging AI-generated insights to define new directions. This transition is likely to result in more orchestrated and decentralized AI agents performing domain-specific functions, such as prioritizing backlog items, balancing resource allocation, or validating compliance rules.
Only about 11% of companies have actually put these systems into the real world [2]. We’re still in the “figuring it out” and “the correct time to implement agentic AI” phase.
The New Program Manager Playbook
Contrary to fears of widespread job displacement, industry leaders increasingly view AI as a complementary force. For example, McKinsey’s recent analysis of workforce shifts highlights growth in client-facing roles and strategic functions, even as automation absorbs routine tasks [3]. Program managers will not disappear; instead, their domain expertise, strategic oversight, and ability to work alongside AI systems will become even more valuable.
To thrive in this evolving environment, program managers and similar leaders must develop new capabilities:
- AI Fluency (Not Coding): Understanding how AI agents reason and operate is essential. When an AI proposes an unexpected or questionable shortcut, leaders need the intuition to recognize why it may be misaligned.
- Data Literacy: AI systems are only as effective as the data they consume. Poor-quality data will lead AI agents to make flawed decisions—often faster and at a larger scale than humans.
- Trust and Governance: Leaders are responsible for ensuring AI systems operate ethically, securely, and reliably, without generating misleading or inaccurate outputs.
- Empathy and Change Leadership: AI adoption often creates uncertainty among teams. Employees do not want to feel replaced by automation. The ability to lead through change, address concerns, and manage the human impact of AI transformation will be a critical leadership strength.
Orchestrating Intelligence, Not Checklists
The “Agentic Revolution” isn’t about replacing the person running the program; it’s about freeing that person to actually lead. We’re moving away from a world where we manage checklists and into one where we orchestrate intelligence. It’s a bit intimidating, sure, but it’s also the first time in a long while that the “future of work” actually looks like it might involve less busywork.
References
[3]