Companies are spending trillions on AI. The C-suite doesn’t know who is in charge of it

A Pearl Meyer survey finds only 34% of C-suite executives say it’s consistently clear who owns corporate AI decision-making, vs. 53% of board members and 57% of senior managers—signaling execution risk. The article cites Gartner projections for AI spending of $2.5T in 2025 (+44% YoY) and $3.3T in 2026, while CEOs also report job-at-risk stakes (80% in the US) tied to AI failures. Net, the message is that AI ambition and spending are outpacing the leadership structure and change-management capacity, raising the odds of a bumpy, outcome-justification phase.

Analysis

The more important signal is not slower AI adoption; it is governance failure. That typically shows up first as budget re-authorization risk: projects keep running, but CFOs and boards force harder hurdle rates, which compresses multiples for application-layer software and systems integrators that sold “transformation” rather than measurable productivity. In the next 1-3 months, the market is likely to reward vendors with auditable workflows, controls, and ROI analytics, while punishing names where AI remains a feature story instead of a line-item savings story.

Second-order, this is a tailwind for change-management, IT advisory, and benchmarking businesses because enterprises will need outside validation before renewing spend. That is structurally supportive for Gartner (IT) over 6-18 months: when internal ownership is fuzzy, boards buy external process and peer-comparison frameworks to de-risk decisions. By contrast, broad enterprise software names with elevated AI expectations can face a “show me” phase into earnings season if management cannot tie pilot activity to operating margin or seat expansion.

The contrarian view is that this is not a capex collapse signal; it is an execution-timing signal. Spending likely stays high, but the mix shifts toward infrastructure and governance, which means the winners are the toll collectors, not the storytellers. The risk to a short-basket thesis is that vendors keep capitalizing on urgency and companies keep layering AI into existing budgets, delaying visible disappointment by a quarter or two; the falsifier is a positive revenue re-acceleration or margin uplift from named AI modules by the next two reporting cycles.

Key Decisions for Investors

• Long IT (Gartner) on pullbacks over the next 1-3 months: governance anxiety should increase demand for advisory/benchmarking; target a 10-15% relative outperformance vs. the software index if AI ROI scrutiny intensifies.
• Short a basket of high-multiple enterprise software names where AI monetization is still anecdotal (e.g., NOW/SNOW/TEAM) into earnings: risk/reward is attractive if guidance does not show margin or ARR lift from AI within 1-2 quarters.
• Pair trade long IT / short a software ETF or broad AI application proxy for 3-6 months: thesis is that spending persists but migrates away from narrative-driven software toward validation and control layers.
• Set an alert on next-quarter guidance for explicit AI revenue or productivity metrics; if management teams quantify savings, cover shorts immediately because the market will re-rate on proof rather than adoption rhetoric.
• If owning AI infrastructure already, reduce upside in names dependent on enterprise software monetization and keep exposure in semis/cloud where spend is less contingent on organizational alignment.