A whopping 3% of boardroom execs spy productivity and costs savings from AI

Gartner reports that while three-quarters of corporate board members expect AI to raise productivity and cut costs, only 3% of organizations have met their targets; 38% are over budget, and 85% still plan to increase spending next year. Only about one-fifth of employees’ time is freed up by AI, while Forrester expects AI to account for 20% of business technology spending next year and says more than half of CIOs are likely to miss ROI targets. Apollo Global Management says AI productivity gains have so far accrued to the “Magnificent Seven,” with no sign of a boost across the rest of the S&P 500.

Analysis

The investable distinction is between selling AI capacity and capturing customer productivity. Persistent budgets can support near-term revenue for infrastructure and platform providers, but they do not yet validate broad enterprise software monetization or justify assuming economy-wide margin expansion. If freed employee time is redirected into new work rather than removed from payroll, the first-order payoff is potentially more output or growth—not near-term labor-cost savings. Training and workflow redesign may therefore absorb part of the return and delay margin evidence.

Over the next 1–3 months, watch earnings commentary for AI-related revenue conversion, customer budget scrutiny, and whether spending is concentrated among a few large platforms. Over 6–18 months, the key test is measurable improvement in output per employee, operating margins, or revenue per employee outside the largest technology companies. A reversal in CIO budgets or a shift from pilots to cancellations would expose vendors whose demand depends on continued experimentation. Conversely, sustained adoption accompanied by productivity and margin gains would challenge the cautious thesis. Treat survey results as directional, not proof of company-level ROI.

Key Decisions for Investors

• Avoid treating broad corporate AI spend as evidence of broad AI earnings growth. Separate infrastructure/platform exposure from application vendors and end-user companies; require disclosed customer conversion or measurable productivity before paying for an ROI narrative.
• Relative-value watch: consider long QQQ versus short RSP only as a tactical expression of near-term earnings concentration in large technology platforms, not as a clean AI trade. Size modestly; unwind if platform earnings fail to convert AI investment into revenue or if equal-weight earnings breadth improves.
• Do not underwrite immediate labor-cost savings across enterprise software customers. For application vendors, monitor renewal rates, seat growth, net retention, and customer commentary on budget scrutiny; weakening metrics would be a reason to reduce exposure rather than assume AI spend offsets pressure.
• Catalyst check for the next 1–3 months: track major platform capex and AI revenue disclosures alongside enterprise customer ROI evidence. If spending rises without improving productivity or margins over subsequent quarters, expect a higher risk of budget rationalization and valuation compression.