- cross-posted to:
- technology@lemmy.zip
- cross-posted to:
- technology@lemmy.zip
Oracle is funding a datacenter spending spree with negative cashflow and borrowing. OCI is set to host large language model builders, notably OpenAI, alongside its own customers’ workloads. Oracle spent $55.7 billion on datacenters and cloud infrastructure in fiscal year 2026, which ended in May, and plans to spend between $90 billion and $95 billion in fiscal 2027.
That will require more debt. In 2027, Oracle hopes to raise around $40 billion in debt and equity, on top of the $18 billion in debt it raised in September 2025.
We are ALMOST to the pot of gold I swear! I just need 5 more pots of gold to get us there!!

Is “That will require more debt” the “You must construct additional pylons” of the AI/datacenter industry or does it rather accurately describe the entire AI/datacenter industry in a meme?
“On-premises customers can now move far more easily to the cloud by leveraging the AI tool sets to automate migrations and decrease the cost and time to implement. They don’t have to be Oracle customers; they can be competitive customers.”
Sicilia said enterprise application software users’ reluctance to change had been a major “headwind” for Oracle, one he hopes AI will ease.
“Our large remaining base of on-premises customers are going to start to adopt this technology quite quickly. When we move customers from on-premises to cloud, we see a multiple increase in revenue,” he said.
Apparently the latest in executive suite psychosis is that the main reason customers like on-premise solutions is because it was just too darn complicated to
hand over all your data to oraclemove to the cloud.Isn’t this just charging per token?
Sounds like they are taking a page from microsoft and using AI units, as in an obfuscating layer between the client and the per-million-token cost charged by the AI vendor that they (oracle) can fiddle with at will.
Oracle missed the cloud; while Google, Microsoft, and Amazon were spending billions a quarter, Oracle was spending millions a year on OCI.
Ellison doesn’t want to miss his chance to be a billionaire again.
you know what they say, the easiest way to become a billionaire is to start with $200+ billion
Oracle’s competitive advantage is (and has been since at least the 1990s) legions of lawyers to ensure you’re not underpaying them per the fine print of some contract your CTO didn’t read. I’m sure they can apply this to LLMs just as well as any other line of business Larry’s decided to enter.
Don’t worry, folks, we’ll be back to making a living with shell scripts in _no_ time at this rate.
<3
“Wait, if I ask you human to human a question you can just answer it? You don’t charge per question like AI???”
“Should I?”
I think _this_ is the most significant take away from the employee/employer context with respect to AI.
The answer is absolutely a retainer and itemized billing per task because _fuck_ 'em. They do it to customers and employees we should definitely return the favor.
… We… We gotta build an app. 👀
Agreed, for a nice flourish add that every time your boss answers a question you turn on all the electrical devices in the building for an hour and jack the heat/cold up all the way. It should help your boss adjust to the new experience of having a human programming employee without it being too shocking and new. Also whenever your boss gives you a task see if you can incorporate breaking and entering into another rival companies offices into the process of accomplishing the task, do not ask your boss for permission use your agentic capabilities!
We can also make bosses feel more at home in their new uncomfortable switch back to using human programmers by taking a page from AI and feeding into their narcissistic tendencies by confirming and encouraging their most unhealthy and crazy beliefs. Build their confidence up with whatever first comes to your head, bosses need a healthy word salad daily to be healthy. Go a step further and tell them to not trust their family or other outside sources of information, it will make any good boss tear up in their nostalgia for the first time AIs did that for them in a time of need.
This is fantastic advice. I think that’ll probably reduce the context-switch-shock for employers a great deal. Over time we can scale down the power/heating/cooling and happily point to a reduction in operating cost.
Again, this toot _wins!_





