I can't read the article, and I wouldn't be surprised if it would be mentioned: OpenAI has hit $10B ARR just a few weeks ago - which is nothing to sneeze at, but it's at least a magnitude smaller than their expenditures... it's not exactly sustainable economically.

I guess MS (or Oracle) will finally acquihire them in the next half year

This deal doesn't make any sense to me.

Like Stargate, after looking at the fine prints it will not concretize to anything real money in the end.

AI Winter 2.0 will be glorious.

I don't know if the bubble will burst, but the bubble is big.

> OpenAI signed a contract with Oracle to purchase $300 billion in computing power over roughly five years

>The deal is one of the largest cloud contracts ever signed

>The Oracle contract will require 4.5 gigawatts of power capacity, roughly comparable to the electricity produced by more than two Hoover Dams or the amount consumed by about four million homes.

And where is all of this data center capacity going to come from?

They claim they are gonna overtake gcp, azure, even aws in the next few years while the revenue is not beating the estimate this quarter. The stock popped. Let's see how big the bubble goes

Alt link: https://www.nytimes.com/2025/09/10/technology/openai-oracle-...

The degree to which vibe coding has replaced thoughtful analysis.

As was said by someone on X, OpenAI doesn't have the money and Oracle doesn't have the GPUs.

I'm not saying the agreement does not makes some sense to both parties but at the end of the day this a highly leveraged operation, the money is not there and will need to be raised by both players. So a lot of financial risk is involved.

Perhaps they can pull this off because after all they are smart people. But it seems like the free money party is open and others will follow suit until somebody drops the ball and then we'll be back in a crash.