There is a growing mismatch building inside enterprise IT budgets. The invoice arrives, the line items are familiar — seat licenses, platform fees, per-ticket costs — but the work those charges reflect is no longer being done by the people those seats represent. It is being done by AI agents, autonomously, end-to-end, before a human enters the loop.
This is not a future scenario. It is happening now, at scale, across industries.
Early recognition and strategic investment will give organizations a massive advantage in the next decade. At Automation Anywhere, my team has drawn new data from more than 70 enterprise deployments as part of our ITSM report. Our findings show that our AI-powered IT service agents are resolving up to 80% of employee support requests autonomously. Call volumes are dropping by up to 50%. Issues are being prevented before they ever become tickets at twice the previous rate. And ITSM licensing costs are falling by as much as 53%, with some large organizations realizing more than $5 million in annual savings.
These are structural shifts that raise a question on SaaS economics that every CIO should be wrestling with today: If AI is doing the work, why are we still paying for software as if humans are?
Software Can Now Do the Work It Was Designed to Support
For decades, enterprise software pricing made intuitive sense. More employees meant more licenses. More tickets meant more costs. The logic was built on a foundational assumption that humans were the primary unit of work, and software existed to make them more efficient.
That assumption is breaking down.
This dynamic has a name: the ‘Illusion of Technology Servitude’ — the deeply conditioned belief that organizations must adapt to their technology, rather than designing technology that adapts to them.
AI agents don’t use seats or generate tickets in the traditional sense. They resolve issues directly, and at a scale and speed that no service desk team could match. When software is no longer a tool that assists the worker but a system that performs the work itself, the entire logic of how that software is priced begins to crack.
That is what makes the license-per-seat model increasingly misaligned with enterprise reality. It was built for a world where scale meant adding people, and adding people meant adding software access. But when AI agents can absorb more of the work without a corresponding increase in human users, growth should not automatically trigger another round of seat expansion. The pricing model has to move closer to the work being performed, the outcomes being delivered, and the value being created.
Also Read: CIO Influence Interview with Hugo Dozois-Caouette, CTO and Co-founder at MaintainX
The Economics Are Changing Faster Than the Contracts
CIOs are already navigating what Gartner has described as a roughly 9% annual SaaS “inflation tax” with rising costs across an increasingly fragmented software portfolio. At the same time, Gartner projects the global ITSM market to roughly double by the end of the decade and predicts that 40% of enterprise applications will be integrated with task-specific AI agents by the end of 2026, up from less than 5% in 2025. 1].
AI is reducing the volume of human-mediated interactions that traditional pricing models were built to monetize. Organizations that fail to recognize this will find themselves locked into commercial models that no longer reflect how work actually gets done.
AI has fundamentally made the license-per-seat model obsolete.
IT Service Operations Are the First Proving Ground — Not the Last
ITSM is the canary in the coal mine. It generates high volumes of repeatable, structured requests. Costs are measurable. Outcomes are visible. When AI agents resolve the majority of support requests before a human is involved, the economic impact is immediate and attributable.
This isn’t limited to IT either. Early deployments are expanding into HR service desks, employee experience workflows, and customer-facing operations. The pricing pressure visible in ITSM will inevitably spread to any enterprise software category whose commercial model assumes large numbers of humans interact with workflow tools on a regular basis.
This is the other half of the story that tends to get lost in conversations about AI and cost reduction. When AI handles the routine, human IT professionals are freed to operate at a higher altitude, applying judgment, context, and strategic thinking that no autonomous system can replicate. The question is not what AI can do. It is why certain work should remain human-only when it doesn’t require human judgment.
What This Means for CIOs Right Now
The shift to an AI-driven ITSM strategy requires a straightforward, though organizationally challenging, change. Before renewing legacy ITSM software, assess where AI is already automating work and ensure your pricing reflects current reality, not the past. Redefine success: abandon ticket closure and seat utilization metrics. Focus instead on resolution quality, issue prevention, cost-per-outcome, and employee experience.
This is an operating model conversation, and it belongs at the CIO level.
The Clock Is Already Running
Enterprise software has historically been designed to help people do work. What we are beginning to see now is software that performs the work itself. Every assumption about how software is priced, valued, and measured has to change.
All CIOs and tech leaders need to ask themselves this question: Is your organization positioned to lead through this shift, or will you have to explain during next year’s budget review why the bill still doesn’t match the work?
Source [1] Gartner, 5 Ways SaaS Vendors Are Increasing Costs and What to Do About It, Hannah Decker, James Smith, 23 October 2025
About The Author Of This Article
Mihir Shukla is CEO and Co-Founder at Automation Anywhere
About Automation Anywhere
Automation Anywhere is a premier enterprise software platform that combines Robotic Process Automation (RPA), goal-driven AI agents, and workflow orchestration. Headquartered in San Jose, CA, it empowers organizations to automate complex, end-to-end business operations without needing extensive coding.
Catch more CIO Insights: What Does “Job-Ready” Really Mean in IT and Cybersecurity?
[To share your insights with us, please write to psen@itechseries.com ]
