On July 1, 2026, Gartner stated that from that point through 2030, up to about US$234 billion in enterprise application software spending will be exposed to "agentic arbitrage," representing roughly 20% of enterprise application SaaS spending in 2030. This is a risk exposure in which spending may be reallocated; it does not mean that 20% of SaaS will inevitably disappear or be replaced.
Gartner defines agentic arbitrage as AI agents completing tasks across multiple systems, reducing the need for users to operate multiple traditional software interfaces directly. The impact centers on how software value is priced and delivered: buyers may shift from seat counts, interfaces, and feature counts toward process outcomes, cross-system orchestration, and measurable return on investment. Gartner also emphasizes that SaaS is not being destroyed, but is undergoing a transformation in business models and delivery methods.
CIOs should not terminate contracts early based on a forecast alone. A more verifiable approach is to first list the tasks and workflows each SaaS product actually supports, then check whether AI agents can access the required data and APIs, preserve business context and audit trails, and roll back safely on failure. Only when cost, service levels, data portability, and accountability boundaries can all be compared is it possible to decide whether to renew, renegotiate, use a hybrid approach, or replace.
Unima recommends organizing a software portfolio review into four columns: tasks and usage; data and integration dependencies; contracts and outcome metrics; and exit and rollback testing. Start by picking one reversible, low-risk process for a controlled comparison, and do not assume, without validation, that AI agents will necessarily be cheaper or more effective. This is a practical checklist Unima compiled from Gartner's material and does not represent specific client results.
Official source: Gartner Newsroom, "Gartner Says $234 Billion in Enterprise Application Software Spend Is at Risk from Agentic AI," 2026-07-01, https://www.gartner.com/en/newsroom/press-releases/2026-07-01-gartner-says-us-dollars-234-billion-in-enterprise-application-software-spend-is-at-risk-from-agentic-artificial-intelligence. This article was fact-checked on 2026-08-24. Gartner's forecast describes market risk exposure and does not mean every enterprise will replace SaaS at the same pace or in the same way.