Topic / Subject
Reuters reports a source says OpenAI is targeting roughly $600 billion in total compute spend through 2030 as it lays groundwork that could support a future IPO story.

TL;DR
That’s an absurdly big number — even for AI. If accurate, it explains why fundraising, chips, and data-center power have become the real battleground.

Key Details

• Per Reuters, a source familiar with the matter said OpenAI is targeting about $600B in total compute spend through 2030.
• Per Reuters, the source said OpenAI’s 2025 revenue totaled $13B and it spent $8B during the year.
• Per Reuters, a fundraising round seeking more than $100B could value OpenAI at about $830B, and Reuters reports Nvidia is nearing a $30B investment as part of that round.
• Per Reuters (citing CNBC), OpenAI expects more than $280B in total revenue by 2030, split nearly evenly between consumer and enterprise.
• Reuters also notes prior talk from Sam Altman about a much larger infrastructure ambition, and reports The Information said inference costs rose sharply in 2025.

Breakdown
If you’ve ever wondered why AI feels like an arms race, this is the number that makes it click.

Per Reuters, OpenAI’s compute spend target through 2030 is around $600B, based on a source familiar with the matter. That’s not “cloud bill growth” — that’s “build an empire” spending, and it puts pressure on everything upstream: chip supply, data centers, and power availability.

Reuters also frames it in a bigger money loop: revenue/spend in 2025, massive projected revenue by 2030 (per CNBC via Reuters), and a fundraising round so large it reads like a sovereign-budget headline. Even if these figures shift, the direction is clear: OpenAI is planning at a scale where infrastructure becomes strategy.

The IPO angle is the subtext: companies don’t talk like this unless they’re building a narrative for investors (even if timing and valuation are still moving targets).

What to Watch Next

• Whether the reported fundraising round actually closes and at what terms
• Any confirmation or updated guidance on compute plans and infrastructure partners
• Data-center and energy constraints that could cap or delay the roadmap
• Any clearer signals on IPO planning (timing, structure, or milestones)

Sources

• Reuters — OpenAI expects compute spend of around $600 billion through 2030, source says
• The Indian Express — OpenAI expects compute spend of around $600 billion through 2030: Reports

Comment
If OpenAI really spends at this scale, what becomes the bigger bottleneck first: chips, electricity, or data-center construction?