Meta Platforms has disclosed USD 825 billion in future lease and contractual commitments, highlighting the enormous financial obligations building up as the Facebook and Instagram owner expands its artificial intelligence (AI) infrastructure and prepares for a new phase of competition in the technology industry.
The commitments, detailed in a quarterly filing with the US Securities and Exchange Commission (SEC), include agreements covering data centres, cloud computing capacity, servers, networking equipment and energy supplies.
The disclosure comes as investors increasingly question how quickly technology companies can turn their vast AI investments into sustainable returns.
The total comprises approximately USD 355.73 billion in leases that have not yet commenced and USD 454.52 billion in non-cancellable contractual commitments.
Meta also disclosed a contingent obligation of up to USD 14.72 billion to purchase cloud capacity over five years, taking the combined potential exposure to about USD 825 billion.
These figures represent obligations extending across multiple years, rather than expenditure due immediately.
The leases cover infrastructure that will begin operating between the remainder of 2026 and 2036, with terms ranging from more than one year to as long as 30 years.
The non-cancellable contracts largely relate to third-party cloud capacity, servers and network infrastructure, energy service agreements for data centres, and consumer hardware developed by Reality Labs, Meta’s virtual and augmented reality division.
Of the contractual commitments, about USD 45.85 billion is due in 2026 and USD 121.41 billion in 2027, illustrating how a substantial portion of the obligations will fall due over the next two years even as longer-term infrastructure agreements continue to accumulate.
The contingent cloud-capacity obligation is different from the company’s fixed commitments.
Meta said the amount could decrease if the provider sells the relevant capacity to other customers, meaning the full sum may not ultimately become payable.
The latest disclosure marks a substantial increase from July, when Meta reported nearly USD 700 billion in future lease and contractual commitments.
The rise reflects the scale and speed of the infrastructure build-out as the company seeks additional computing power for its AI models, products and services.
Meta is also increasing its near-term capital expenditure.
The company narrowed its 2026 capital spending forecast to between USD 135 billion and USD 145 billion, including principal payments on finance leases, raising the lower end of its previous range of USD 130 billion to USD 145 billion.
Capital expenditure reached USD 34.91 billion in the third quarter, underscoring the resources required to expand computing capacity.
Meta has been investing in data centres, specialised chips, networking equipment and power infrastructure to support AI development and deployment.
The company aims to bolster its competitive edge against rivals like Microsoft, Alphabet, and Amazon, who are also investing heavily in AI infrastructure. Access to advanced computing capacity has become a strategic priority as technology companies race to develop increasingly capable models and integrate them into consumer and enterprise products.
Chief executive Mark Zuckerberg has positioned AI as central to Meta’s future, arguing that investments in advanced models can improve advertising, content recommendations and user engagement while creating new products and services. The company’s advertising business remains its principal source of revenue and provides the financial foundation for its expansion.
Meta’s third-quarter revenue rose 26% year on year to USD 64.60 billion, supported by growth in advertising. However, operating income fell 37% to USD 12.99 billion, while total costs and expenses increased 68% to USD 51.61 billion.
The figures included a USD 9.89 billion charge related to an agreement resolving litigation brought by US state attorneys general over youth mental health and social media.
Net income increased to USD 9.86 billion from USD 2.71 billion a year earlier, although the comparison was affected by a substantial tax provision in the prior-year period. Meta also reported a negative free cash flow of USD 2.12 billion for the third quarter, despite generating USD 32.79 billion in cash from operating activities.
The figures illustrate the tension between a profitable core business and the capital demands of building an AI platform.
Investment in advertising improvements can be financed, but infrastructure spending also creates long-term obligations that future cash generation must support.
Meta has said its AI investments are already producing incremental advertising revenue through better recommendations, stronger engagement and more effective monetisation. It expects the technology to create further opportunities, although newer AI products may take longer to generate meaningful returns than improvements to its established advertising business.
The company has also earmarked USD 10.80 billion in money market funds as restricted cash equivalents to meet escrow requirements under certain multi-year infrastructure purchase agreements.
The funds are expected to be released between 2028 and 2030 as the underlying obligations are fulfilled.
For investors, the issue is not simply the size of Meta’s commitments but whether the spending will generate sufficient returns to justify the financial exposure. Long-term leases can secure access to essential infrastructure, while non-cancellable contracts provide capacity certainty.
They can also reduce flexibility if demand for computing power, energy costs or the economics of AI services change.
The commitments are not equivalent to debt, and their disclosure does not mean Meta must pay the entire amount at once.
Nevertheless, they demonstrate how AI is changing the financial profile of major technology companies, shifting a growing share of investment towards physical infrastructure and long-duration contractual obligations.
Meta’s challenge will be to sustain growth in its advertising business while converting AI spending into commercially valuable products.