Microsoft is down 13% this year, while the iShares Expanded Tech-Software Sector ETF has fallen nearly 20%, as investors worry AI agents could reduce software demand. Microsoft executive Rajesh Jha argues the opposite: more AI agents could create more per-seat license opportunities and support revenue growth. The article frames the pullback as a potential buying opportunity, with Microsoft now trading at 26x trailing earnings versus nearly 40x last year.
The market is likely pricing AI as a net destroyer of software seats, but the more important second-order effect is credential proliferation: if workflows get decomposed into multiple agents, software usage can scale faster than headcount falls. That is structurally supportive for the large platform vendors with identity, admin, and compliance layers embedded in enterprise workflows, because every incremental agent creates an authentication, audit, and provisioning event that is monetizable. In other words, the revenue risk is less about fewer humans and more about whether incumbents can charge for machine users at the same or higher effective rate. The biggest winner is the software layer that sits closest to enterprise control points, while the biggest losers are point solutions with weak switching costs and pricing tied tightly to named human seats. If agent deployment is real, the first budgeting response from CIOs will be to rationalize licenses, but the second response will be to expand spend on security, workflow orchestration, and identity management to govern AI permissions. That should favor MSFT and adjacent infrastructure beneficiaries over narrow application vendors, and it also creates an indirect tailwind for compute suppliers like NVDA as inference demand rises with every additional autonomous workflow. The contrarian read is that the selloff may be too broad because the market is discounting only seat compression and ignoring software consumption expansion. But the timing matters: near-term multiples can remain pressured for several quarters while customers pilot agents and renegotiate contracts, so the upside is more likely to show up in FY26/FY27 than in the next earnings cycle. A real reversal would require evidence that agent adoption is driving net-new software usage faster than seat attrition, rather than merely substituting lower-cost human labor. From a positioning standpoint, this is a better relative-value than absolute-long setup: the risk/reward is strongest where pricing power and workflow ownership are already entrenched. The cleanest expression is to own the platforms that can repackage access around identities, agents, and governance, while fading the basket of legacy software names still exposed to named-user contraction. If the thesis is right, the market should eventually re-rate software from "per-seat" to "per-workflow," but that shift will likely be uneven and winner-takes-most.
AI-powered research, real-time alerts, and portfolio analytics for institutional investors.
Request DemoOverall Sentiment
mildly positive
Sentiment Score
0.15
Ticker Sentiment
