Meta outlook revised downward by S&P on cash flow and capex concerns

S&P Global Ratings changed Meta’s credit outlook, not its current credit rating. Meta remains rated AA-, but the outlook moved from stable to negative. The change signals that S&P sees increased uncertainty around Meta’s financial direction, especially its rising capital spending and negative free cash flow.

An AA- rating indicates very strong credit quality. S&P’s highest rating is AAA, while ratings below BBB are considered junk bonds. “Negative” does not mean Meta has already been downgraded. It means S&P is less confident that the company’s financial profile will remain strong enough to support the existing rating.

The concern centers on Meta’s planned infrastructure spending and unclear AI returns. S&P questioned whether Meta can monetize its AI investments as effectively as peers. It also said Meta may not reduce spending as previously expected, even though capex growth could begin flattening in 2028.

Meta expects to spend $135 billion to $145 billion on capital projects during the full year. This spending is mainly tied to the infrastructure needed for its AI initiatives, including data center buildouts. The scale matters because it is large enough to push the company’s free cash flow into negative territory.

Meta said it plans to raise $95 billion to help fund its AI efforts. The financing plan starts with $35 billion in underwritten public equity offerings. Meta then plans to raise another $60 billion through an at-the-market program, which sells shares over time.

The figures show how Meta is trying to finance an unusually expensive expansion. The company posted negative free cash flow of $2.12 billion in the third quarter, its first move into the red. S&P remains uncertain about how quickly AI revenue will cover these growing costs.

Higher-than-expected revenue did not resolve investors’ larger concern: Meta is spending heavily while generating less available cash. The company posted negative free cash flow of $2.12 billion in the third quarter, entering the red for the first time. That result suggested that current cash generation was not covering its investment needs.

Meta also expects full-year capex of $135 billion to $145 billion. Investors therefore had to weigh a positive revenue result against a much larger infrastructure bill. S&P’s negative outlook added another warning, questioning whether Meta can turn AI investment into revenue that exceeds its rising costs.

The market reaction was immediate. Meta shares fell 7% after the earnings report, despite the revenue beat. The stock finished October down 15%, its worst month in four years. Those moves show that investors were focused more on cash flow, spending discipline, and AI monetization than on revenue alone.

Free cash flow is the money a company has left after paying for normal operations and capital investments. It matters because this cash can support expansion, repay debt, return money to shareholders, or provide a cushion during weaker periods. Positive free cash flow generally gives a company more financial flexibility.

Meta reported negative free cash flow of $2.12 billion in the third quarter. That means its cash needs, including major infrastructure investments, exceeded the cash generated during the period. Meta said this was its first time going into the red on free cash flow.

Negative cash flow is not automatically a crisis when a company is investing for growth. However, it becomes more important when spending is exceptionally large and the timing of returns is unclear. S&P questioned when Meta’s AI products will generate enough revenue to exceed rising costs. Until then, Meta may need outside financing or tighter spending control.

S&P’s concern is a timing and scale problem. Meta is spending heavily today on computing infrastructure, data centers, AI agents, and foundation models. The agency is not convinced that the resulting revenue will become material soon enough, or grow large enough, to exceed those costs over the next few years.

Meta recently debuted its Muse AI agent app and related foundation models. However, S&P said it remains unclear when these revenue streams will make a material contribution. Meanwhile, Meta expects full-year capex of $135 billion to $145 billion and has already reported negative free cash flow.

The uncertainty is not unique to Meta. S&P said peers are also building data centers aggressively and going cash flow negative. Still, the agency questioned Meta’s ability to monetize AI relative to those companies. It expects capex growth to flatten beginning in 2028, but is less certain Meta will actually curtail spending as previously expected.

Raising $95 billion could give Meta substantial cash to continue building AI infrastructure and funding related products. The stated plan relies on equity: $35 billion through underwritten public offerings and $60 billion through an at-the-market program. Selling new shares can dilute existing shareholders because ownership is spread across more shares.

The article does not describe a new debt-financing plan. Therefore, this specific $95 billion proposal would not itself raise Meta’s debt levels in the way borrowing would. However, raising equity can still affect shareholders by increasing the share supply and potentially changing how investors value the company.

The proceeds could help Meta keep funding data centers and AI initiatives even while free cash flow is negative. They would also reduce immediate pressure to slow construction. At the same time, the financing need highlights S&P’s concern that spending may stay elevated while the timing of meaningful AI revenue remains unclear.

Companies turn AI infrastructure into profits by using it to sell valuable products or make existing businesses more effective. Common routes include advertising tools, paid subscriptions, usage-based access, enterprise services, and licensing. The basic goal is to convert expensive computing capacity and AI models into recurring revenue.

The mechanism depends on scale and efficiency. A company builds data centers and trains models, then serves many users or business customers from that investment. Revenue can rise as adoption grows, while the cost per request may fall through better hardware use, software improvements, and model efficiency. Profit appears when total revenue exceeds infrastructure, research, staffing, and operating costs.

The article does not specify Meta’s exact AI revenue model. It identifies the Muse AI agent app and related foundation models, but says it is unclear when they will make material contributions above rising costs. That uncertainty is central to S&P’s negative outlook and Meta’s funding plans.