Blab Business
Meta Raises 2026 Capital-Expenditure Forecast Amid Rapid AI Build-Out
Meta Platforms, Inc. announced Thursday that it will spend between $130 billion and $145 billion on capital expenditures in 2026, up from a prior range that began at $125 billion. The company said the increase raises the lower bound of the forecast rather than the upper bound, signaling a commitment to a larger AI infrastructure spend.
The guidance follows a strong second‑quarter performance. Meta reported revenue of $60.8 billion, a 28 % year‑over‑year jump that marks the company’s fastest growth since late 2021. The lift was driven largely by the advertising business, which Meta said benefited from AI‑enhanced recommendation algorithms that kept users engaged on Facebook and Instagram. The advertising engine was described as the primary source of the revenue surge.
However, the quarter also highlighted a sharp decline in cash generation. Free cash flow fell to $784 million, down 91 % from $8.55 billion a year earlier. Meta attributed the drop to the scale of its AI build‑out, which is consuming the cash generated by the advertising engine. Earnings per share were $6.18, missing analysts’ expectations of $7.22, and were impacted by a $2.4 billion legal charge that Meta recorded in the quarter.
The legal charge is part of a broader litigation context. A group of U.S. states is seeking up to $1.4 trillion in penalties over claims that Meta’s products are designed to addict young users. Meta said the litigation shadows every quarter and that the company is addressing the risks.
User growth remained steady. Meta reported 3.6 billion daily active users across its apps, up 3 % year‑over‑year, after a dip in the previous quarter. The company said the user base expansion provides a foundation for its AI initiatives.
During the earnings call, CEO Mark Zuckerberg explained how the capital spend will be used. He said a large share of compute will train Meta’s AI models, feed the core advertising business, and power “personal agents” that he expects to become a major consumer product. Zuckerberg also outlined a second‑act strategy: Meta plans to grow a large business serving large customers by renting out AI compute. The company said the same data centers that are currently a cost will eventually pay their own way.
Meta’s guidance reflects a broader trend among Big Tech. The company’s capex plan is similar to those of Amazon, Alphabet, and Microsoft, all of which are investing heavily in AI infrastructure. Meta’s approach is distinctive in that it is positioning itself as a cloud‑style provider without yet being one, and it is tying the investment to a superintelligence ambition.
Investors have largely accepted the trade‑off. Meta’s shares have largely held up through the spending spree, on the bet that a company growing revenue at 28 % can afford to invest heavily, provided the growth does not stall. The company’s next earnings report will provide more detail on how the AI build‑out is affecting profitability and whether the cash‑flow squeeze will ease.
The key unresolved question is timing. Meta can afford the current level of spending, but the collapse in free cash flow raises questions about how many more quarters the company can sustain the pace before the return on investment materializes.
In summary, Meta’s 2026 capital‑expenditure forecast signals a deepening commitment to AI infrastructure, even as it faces a sharp decline in free cash flow and ongoing legal challenges. The company’s strategy to monetize its data centers through compute rentals and personal agents remains to be proven, but the market has so far accepted the risk profile.
The guidance follows a strong second‑quarter performance. Meta reported revenue of $60.8 billion, a 28 % year‑over‑year jump that marks the company’s fastest growth since late 2021. The lift was driven largely by the advertising business, which Meta said benefited from AI‑enhanced recommendation algorithms that kept users engaged on Facebook and Instagram. The advertising engine was described as the primary source of the revenue surge.
However, the quarter also highlighted a sharp decline in cash generation. Free cash flow fell to $784 million, down 91 % from $8.55 billion a year earlier. Meta attributed the drop to the scale of its AI build‑out, which is consuming the cash generated by the advertising engine. Earnings per share were $6.18, missing analysts’ expectations of $7.22, and were impacted by a $2.4 billion legal charge that Meta recorded in the quarter.
The legal charge is part of a broader litigation context. A group of U.S. states is seeking up to $1.4 trillion in penalties over claims that Meta’s products are designed to addict young users. Meta said the litigation shadows every quarter and that the company is addressing the risks.
User growth remained steady. Meta reported 3.6 billion daily active users across its apps, up 3 % year‑over‑year, after a dip in the previous quarter. The company said the user base expansion provides a foundation for its AI initiatives.
During the earnings call, CEO Mark Zuckerberg explained how the capital spend will be used. He said a large share of compute will train Meta’s AI models, feed the core advertising business, and power “personal agents” that he expects to become a major consumer product. Zuckerberg also outlined a second‑act strategy: Meta plans to grow a large business serving large customers by renting out AI compute. The company said the same data centers that are currently a cost will eventually pay their own way.
Meta’s guidance reflects a broader trend among Big Tech. The company’s capex plan is similar to those of Amazon, Alphabet, and Microsoft, all of which are investing heavily in AI infrastructure. Meta’s approach is distinctive in that it is positioning itself as a cloud‑style provider without yet being one, and it is tying the investment to a superintelligence ambition.
Investors have largely accepted the trade‑off. Meta’s shares have largely held up through the spending spree, on the bet that a company growing revenue at 28 % can afford to invest heavily, provided the growth does not stall. The company’s next earnings report will provide more detail on how the AI build‑out is affecting profitability and whether the cash‑flow squeeze will ease.
The key unresolved question is timing. Meta can afford the current level of spending, but the collapse in free cash flow raises questions about how many more quarters the company can sustain the pace before the return on investment materializes.
In summary, Meta’s 2026 capital‑expenditure forecast signals a deepening commitment to AI infrastructure, even as it faces a sharp decline in free cash flow and ongoing legal challenges. The company’s strategy to monetize its data centers through compute rentals and personal agents remains to be proven, but the market has so far accepted the risk profile.
