OpenAI Lowers Infrastructure Investment Target
OpenAI has revised its infrastructure investment goal downward to $600 billion by 2030, a significant reduction from previously suggested figures. The adjustment reportedly follows pushback from investors and reflects a more cautious strategy regarding the high costs and risks of developing advanced AI. This move is seen as a signal for the broader AI industry to balance rapid growth with fiscal discipline.
- The previous infrastructure investment figure, floated by CEO Sam Altman, was a staggering $1.4 trillion, which has now been reduced by 57%. This revision aims to more closely align the company's capital expenditures with its revenue projections. - The adjustment follows concerns from investors that the company's ambitious expansion plans were growing faster than its potential revenue. OpenAI's spending on running its AI models, known as inference costs, quadrupled in 2025, causing its adjusted gross margin to fall. - To fund its infrastructure, OpenAI is in the process of finalizing a massive funding round expected to exceed $100 billion. This round involves strategic investors, with Nvidia reportedly in talks to invest up to $30 billion. - Other significant investors expected to participate in the funding round include SoftBank, Amazon, and Microsoft. The round could result in a pre-money valuation of OpenAI at $730 billion. - The company projects it will achieve annual revenue of over $280 billion by 2030, with the revenue stream split almost equally between its consumer and enterprise offerings. In 2025, OpenAI generated $13.1 billion in revenue, exceeding its $10 billion target, while spending approximately $8 billion. - This investment is slated to support projects like "Stargate," a multi-year, $500 billion venture with partners including SoftBank and Oracle to build A.I. data centers across the United States and other countries. - Previously, CEO Sam Altman had discussed an even more ambitious, and reportedly separate, initiative to raise between $5 trillion and $7 trillion to overhaul the global semiconductor industry and solve the scarcity of AI chips. - The current $600 billion target, while a reduction, is still a massive figure that dwarfs the typical annual capital expenditures of tech giants like Amazon and Microsoft combined.
