For nine months, the market has been running one very simple trade. AI will write the code, answer the phones and process the insurance claims, so any company that makes money selling human hours to do those jobs is roadkill.
Indian IT stocks got hit in February. Call-center giants slid. In June, a Gartner warning about failed AI projects pushed the whole outsourcing complex lower again. Accenture, the biggest name in the group, was down roughly 30% for the year before this morning.
Then this morning Accenture reported record bookings of $84.5 billion for fiscal 2026, guided to 3–6% revenue growth next year, and the stock jumped more than 20%.
The number worth staring at is not the jump. It’s a sentence in how the company now reports. Accenture stopped breaking out “advanced AI” bookings after the first quarter, and management’s explanation was that AI work “is increasingly being incorporated into larger client projects rather than sold as a standalone service.”
Read that slowly. The fear was that AI would replace the project. The evidence is that AI is becoming part of the project.
That changes the question. If AI work is getting folded into big service contracts, the interesting part is no longer whether the services industry survives. It’s who gets paid for the least glamorous piece of the job: cleaning up messy data so a model can use it, and wiring AI into regulated workflows like claims, underwriting and payment audits, where a wrong answer has legal consequences.
Accenture is a $138 billion company growing in the low-to-mid single digits. It can’t be the cleanest way to play that shift. A smaller company with the same exposure and a much faster growth rate would be.
Here is what that company looks like on paper. More than 60% of its revenue now comes from data and AI-led work, and that slice grew 30% last quarter, the fourth straight quarter of acceleration. It has raised full-year guidance at each of its last two reports. It trades at roughly the same earnings multiple as Accenture does after today’s move. And in July it was sitting at a 2026 low, priced as exactly the kind of business AI was supposed to eat.
So which company is it?
This is where the story stops being about a headline and starts being about the numbers underneath it.
