As AI demand grows, OpenAI becomes a major cloud server consumer, developing custom chips and data centers, with its first AI chip launching in partnership with Broadcom, while expanding cloud partnerships with Oracle and Google Cloud to support its infrastructure
OpenAI has significantly increased its projected spending through 2029 to a staggering $115 billion, reflecting its aggressive investments in artificial intelligence infrastructure and computing power.
According to a recent report by The Information, the updated forecast marks a sharp $80 billion rise compared to earlier estimates. The AI research firm—best known for its chatbot ChatGPT—expects to spend more than $8 billion in 2025 alone, surpassing its earlier projection by $1.5 billion.
Custom hardware and strategic partnerships drive spending
With growing demand for AI capabilities, OpenAI has become one of the world’s largest cloud server customers. To reduce dependency and curb costs, the company is reportedly developing custom server chips and building its own data center facilities. Its first AI chip is expected to roll out next year in collaboration with U.S. semiconductor leader Broadcom, according to reports. The chip will be used exclusively for OpenAI’s internal operations.
In addition to hardware development, OpenAI has expanded its cloud partnerships. In July, it announced a major deal with Oracle to build out 4.5 gigawatts of data center capacity—part of its larger “Stargate” initiative, a project valued at up to $500 billion, involving SoftBank Group. The company has also brought Google Cloud onboard to bolster its computing resources.
Financially, the outlook is steep. The company’s cash burn is expected to exceed $17 billion in 2026, followed by $35 billion in 2027 and $45 billion in 2028.
OpenAI has not issued a public statement in response to the report.
