GE Vernova's Selloff Tests the Financing Assumptions Behind AI Demand

GE Vernova fell nearly 12% on October 22, while Vertiv and Eaton also declined, after Meta's quarterly report forced investors to look past the scale of artificial-intelligence spending and ask how that buildout will be financed. The reaction hit companies that sell power-generation, grid and cooling equipment into the data-center boom.

Meta reported third-quarter revenue of $64.6 billion, up 26% from a year earlier, and capital expenditures of $34.9 billion. It also disclosed an $18.59 billion litigation agreement and a related $9.89 billion third-quarter charge. Separately, Meta announced a new class of non-voting stock. That corporate action does not prove an imminent financing need, but it added to a broader debate about how hyperscalers will fund a rapidly expanding infrastructure program.

The market's concern was not that data-center power demand had disappeared. GE Vernova ended the second quarter with a $176 billion backlog after orders rose 88% organically and revenue increased 22%. Those figures show substantial demand for generation and grid equipment well beyond a single customer or project.

The harder question is conversion. A backlog records contracted work, but revenue and cash arrive over time and depend on customer financing, permitting, supply availability and project schedules. If hyperscalers or their development partners delay projects, equipment suppliers can retain long-term demand while still experiencing near-term order timing or cash-flow volatility.

That distinction helps explain why GE Vernova's shares fell much more than the broader market even though the company had not issued a new warning. Investors had priced in years of strong electricity demand from AI facilities. Meta's spending profile reminded them that equipment demand is only as durable as the capital behind each project.

The selloff also should not be read as evidence that Meta cannot finance its plans. Meta remains a large, profitable company, and its revenue continued to grow. The defensible conclusion is narrower: investors raised the discount they apply to suppliers whose valuations depend on an unusually fast infrastructure cycle.

GE Vernova reports its own third-quarter results on October 28. Orders, backlog composition, customer deposits and cash conversion will matter more than another broad statement about AI demand. If those indicators remain strong, the October decline may prove to have been a financing-risk reset rather than a break in the underlying power-equipment cycle.