IREN pushes its target of more than $4bn of annualised AI cloud revenue into January 2027 after a customer asked for changes, and names Microsoft on its call, CNBC reports

IREN, which rents out Nvidia GPU capacity as an AI cloud, said on 5 November that its target of more than $4 billion of annualised run-rate revenue (ARR), previously set for the end of 2026, “is now targeted for January 2027” and “may extend to February 2027”. The cause, in the company's words, is “a customer-requested change and supplier timing”.

The results release does not name the customer. On the earnings call, co-chief executive Daniel Roberts named Microsoft, CNBC reported, saying Microsoft had made a request to reconfigure and upgrade some of the network infrastructure.

In its fiscal first quarter, IREN's total revenue rose 66% from the previous quarter to $228m, and AI cloud revenue rose 182% to $200m. It still made a net loss of $263m. Its contracted revenue backlog grew 22% to $20.3bn.

The release calls ARR “an operating metric, not a U.S. GAAP measure”, and says revenue actually recognised may be materially lower. It describes the target as fully contracted and fully funded.

The delay came on the day the AI sell-off hit the companies that rent out computing power, and the same day Bloomberg reported that Microsoft had declined an option to expand its Texas deal with another such company, Nscale.

• Confirmed IREN's target of more than $4bn of ARR moves from the end of 2026 to January 2027 and may extend to February 2027, after a customer-requested change and supplier timing. IREN Q1 FY27 results, SEC 8-K
• Confirmed Quarterly revenue was $228m, up 66%, AI cloud revenue $200m, up 182%, and the backlog $20.3bn. IREN Q1 FY27 results, SEC 8-K
• Reported Co-CEO Daniel Roberts named Microsoft as the customer that asked for the change, CNBC reported. CNBC