Most headlines on this survey focus on back-office gains.

Most headlines on this survey focus on back-office gains. The real finding is buried in the footnote: most firms running AI tools run them as disconnected point solutions.

That’s the part that matters. If you’re using AI for data reconciliation in one tool, meeting-note synthesis in another, and compliance document review in a third, you haven’t changed your operating model. You’ve added three more systems to reconcile manually. The duplication problem doesn’t go away. It moves one layer up.

I’ve seen this pattern in every conversation with ops leaders over the past year. The firms pulling ahead are not the ones with the most AI tools. They’re the ones that decided where data lives and how it flows before layering AI on top. The foundation question came first. The tooling followed.

The analysts quoted in this piece frame it as “AI as an operating system for the whole firm.” That framing is right. The precise version: AI compounds when the underlying data is connected, and creates new coordination costs when it isn’t. Most firms are still learning that distinction the expensive way.