Notes · Briefing
The AI Spend Brief
The studies telling your board how enterprise AI is going don’t agree — half say it’s working, half say it isn’t, in the same season — because they aren’t measuring the same thing. This is the three-page version that untangles them, so the number you take upstairs survives the room.
01 — Usage is not impact.
The model is a commodity — you and your competitor license the same one this morning. The difference is what the spend is aimed at, and whether anyone can hold it to a number. Even Meta hit this: it made AI usage a performance expectation, its people optimized the metric, and its CTO ended up killing the internal dashboard — because usage is not impact. We’ve spent our own tokens to turn them into board-readable outcomes — red to black. The spend stops being a line you defend and becomes the line the quarter turns on. We are hired to be the exception. And unlike everyone who sold you the last one, we arrive already accountable to a number.
02 — The numbers don’t disagree. The nouns do.
The studies telling you how AI is going don’t agree — because they aren’t measuring the same thing. Deloitte asks whether your best project met the bar you set for it: 74% say yes. BCG asks whether your whole company is producing value: 74% say no. MIT’s famous 95% says organizations and measures tool deployments — its own report puts three different units in one paragraph. The numbers don’t disagree. The nouns do. Noticing that is the job — and it’s precisely what the team that will answer for your result should be the first to notice.
“The hype on LinkedIn says everything has changed, but in our operations, nothing fundamental has shifted. We’re processing some contracts faster, but that’s all that has changed.”
03 — The sources.
Sources — every figure in this brief, linked. All are self-reported executive surveys; none is a measurement. We say so because the distinction is the point.
Cisco AI Readiness Index 2025 — “only one in three organizations (32%) have a process to measure the impact of their AI initiatives.” n=8,039 senior leaders (500+ employees), 30 markets, double-blind, Oct 14 2025. PDF
McKinsey, The State of AI 2025 — “88 percent report regular AI use in at least one business function”; “Thirty-nine percent of respondents attribute any level of EBIT impact to AI, and most of those respondents say that less than 5 percent of their organization’s EBIT is attributable to AI use.” n=1,993, 105 nations, fielded Jun 25–Jul 29 2025. mckinsey.com
PwC 29th Global CEO Survey — “More than half (56%) say their company has seen neither higher revenues nor lower costs from AI, while only one in eight (12%) report both.” n=4,454 CEOs, 95 countries, fielded Sep 30–Nov 10 2025. PDF
IBM Institute for Business Value / Oxford Economics — “only 25% of AI initiatives have delivered expected ROI over the last few years”; 16% scaled enterprise-wide. n=2,000 CEOs, 33 countries, fielded Feb–Apr 2025. newsroom.ibm.com
Deloitte, State of Generative AI in the Enterprise (Wave 4) — “nearly three-quarters (74%) say their most advanced initiative is meeting or exceeding their ROI expectations.” Question asks about “one of your organization’s most advanced (scaled)” initiatives, benchmarked to “your organization’s expectations.” n=2,773, 14 countries, fielded Jul–Sep 2024. PDF
BCG, Where’s the Value in AI? — “The remaining 74% have yet to show tangible value from their use of AI.” A derived score band (stagnating 25% + emerging 49%) on a weighted 30-capability index; inputs self-estimated by executives, weights and cutoff chosen by BCG. n=1,000 CxOs, 59 countries, Oct 2024. PDF
MIT Project NANDA, State of AI in Business 2025 — “95% of organizations are getting zero return.” v0.1, “Preliminary Findings,” not peer-reviewed. Authors’ own caveat: “directionally accurate based on individual interviews rather than official company reporting.” 52 interviews, 300+ public initiatives, 153 survey responses. MIT gates the report behind a request form; the copy in circulation is a mirror. We cite it because it is the number everyone quotes — not because we rely on it.
The number that survives the board starts with one briefing.
Thirty minutes with the person who has answered at that standard — inside Boeing Defense, the Department of Energy, Nike. NDA first, on request.