Issue 1 · July 14 · Issue 3 · July 28 · Issue 4 · August 5 · Issue 5 · August 11 · Issue 6 · August 19 · Issue 7 · August 26 · Issue 8 · September 1 · Issue 9 · September 8 · Current issue
A weekly read on what the AI noise is hiding.
Last week the diagnosis came from MIT: 95% of enterprise AI pilots return zero profit. This week it came from the corner office. PwC asked 4,454 CEOs whether AI paid off yet, and a clear majority said no. The people who signed the checks are now saying the quiet part out loud.
What the AI noise is hiding. Every stat sourced. The hype named as hype. One number worth acting on. Read the full brief below, free, no email required.
Correction added August 28, 2026, after publication. The closing line read “the spend is racing ahead of the plan by a factor of four.” The 94% and the 23% are both ABI Research, the same 490 supply-chain professionals, and the ratio is sound. But the word “spend” imported a separate Accenture figure from a different, all-industry population. It is a ratio of intent to strategy, not of spend to plan. Corrected in place, this note left standing.
Same disease as last week, now admitted at the top. The money is moving. The value isn’t landing. The full brief shows why the gap sits between the model and the org, not inside the model, and what the highest-intent vertical reveals about the fix.
Last week a study said it. This week the CEOs said it themselves.
PwC’s 29th Global CEO Survey asked 4,454 chief executives, across 95 countries, a simple question. Has AI moved your numbers in the last year? A clear majority said no. 56% report no significant financial benefit yet, meaning neither higher revenue nor lower costs. Only 12% report both. (PwC 29th Global CEO Survey, verified.)
Read that again. These aren’t skeptics on the sidelines. These are the people who approved the budgets, hired the consultants, and stood on stage talking about transformation. Now most of them are telling their own surveyor the return hasn’t shown up.
It’s the same signal MIT sent last week. MIT’s “GenAI Divide” study found 95% of enterprise pilots deliver zero measurable P&L impact. One number came from the lab. This one came from the boardroom. They point at the same thing.
The reason isn’t the model. The models work. It’s the coordination around them that never changed. Here’s what the firms confirmed, and what the highest-intent vertical makes impossible to miss.
The consensus didn’t shift this week. It hardened.
BCG’s AI Radar 2026 still splits success into 70% people and change, 20% process and data, 10% algorithm. The algorithm is the smallest slice. The humans around it are the whole job. PwC’s own read agrees from the top down: the CEOs who report real financial returns are about three times more likely to have built the foundations first, the responsible-AI frameworks and the wiring that lets AI run across the business instead of in a corner. (PwC 29th Global CEO Survey.)
So the split inside the PwC data is the tell. 12% get both revenue and cost gains. 56% get neither. The gap between those two groups isn’t a better model. Everyone can buy the same model. The gap is whether the organization around the model actually changed to use it.
That’s the diagnosis every major firm now shares. Adoption is easy. Value is hard. And the hard part lives in the org, not the tool.
To see the gap clearly, look at the function that wants AI most.
Supply chain is the highest-intent vertical in the market right now. Accenture found 85% of executives plan to increase AI spend in 2026, with one in five planning a rise of 20% or more. (Accenture, reported by SupplyChainBrain, February 2026.) ABI Research surveyed 490 supply chain professionals and found 94% plan to use AI or GenAI for decision support within two years. (ABI Research, verified.) The intent is nearly universal. The checkbooks are open.
Then comes the number that stops the story. Gartner surveyed 120 supply chain leaders who had already deployed AI. Only 23% had a formal AI strategy. (Gartner, June 2025, verified.) Most were running project by project, chasing short-term wins with no plan underneath.
Sit with that pairing. Ninety-four percent intent. Twenty-three percent strategy. Within those same 490 supply-chain professionals, intent is running four times ahead of any written strategy to govern it. That’s not a technology gap. Nobody is short on models. It’s an execution gap, the same one PwC’s 56% are living, shown in one vertical in sharp relief.
What actually works is the inverse. The 12% who see returns built the plan first, then bought the tools. The order matters more than the spend.
Every firm named the disease this week. None named the cure.
They agree on the diagnosis. Value isn’t landing. People and coordination are the real bottleneck. Strategy runs behind spend. Good, all of it, and now confirmed by the CEOs themselves.
Then the prescriptions arrive, and they’re all more of the disease. Build a formal strategy. Stand up the governance. Retrain the workforce. Rewire the data. Buy the platform. Run the program. They tell you coordination is the problem, then hand you a bigger coordination project to manage it.
Nobody names the coordination tax as the thing to remove. They treat the 70% people-and-change cost as a bucket you manage, not a structural drag you can cut. And it’s always a program with a start and an end, never an operating model that keeps the org aligned after the consultants leave. The dependency is the product they’re selling.
One more gap, the same one as last week. Every dataset is Fortune-scale. The PwC CEOs run giants. The supply-chain surveys skew enterprise. The prescribed fix doesn’t exist below a nine-figure budget. The $10M to $250M operator, the one who feels the coordination drag hardest and has the least legacy weight to fight, is standing on unclaimed ground.
That’s the wedge. You don’t manage a tax. You remove it. You install an operating model, not a program. And you do it for the mid-market operator nobody is serving.
56%.
Fifty-six percent of CEOs report no significant financial benefit from AI in the past year, per PwC’s survey of 4,454 of them. Not because the models are weak. Because the coordination around them never changed. The AI landed on top of the same meetings, handoffs, and chase-work that slowed the company down before it arrived.
The bottleneck was never the model. It’s the coordination layer wrapped around it. You don’t optimize that layer. You remove it.
Want to know whether your organization sits in the 56% or the 12%? Start by measuring the coordination tax you’re paying right now. That’s what the Agentic Readiness Score does. It’s 33 questions, it’s free, and it shows you exactly where the drag is.
One weekly read on what the AI noise is hiding. Every stat sourced, the hype named as hype, one number worth acting on. No spam, unsubscribe anytime.
Start by measuring the coordination tax you’re paying right now. Free, a few minutes.
Next Tuesday: another scan, same rule. Cut the hype. Show you what the noise is hiding.