Oracle faces $100M annual bill to back Wisconsin datacenter power promises Oracle says that it could face more than $100 million a year in financing costs to guarantee the power commitments behind a nearly 1 GW datacenter campus it is developing in Wisconsin with Vantage and OpenAI. Local regulators have refused to revisit a decision that they say protects existing customers and improves public …
Sounds like a bunch of communists, the USA should invade them
Oracle paid for this administration, it's ridiculous that some bunch of cheese-eating mid-westerner monkeys should be allowed be allowed to get in the way of their Larry God given right to make more money
I wonder how much of David Ellison's purchase of Paramount and WB has been financed with Oracle shares? If Oracle craters, it might take down their whole empire. If that happened I might have to dig out my tiny violin.
At first, I was a bit confused when I began to see data center "capacity" measured in gigawatts. I assumed I was missing something and that wattage somehow was related to useful computation ability.
But, no, we're denominating these things in the amount of power they consume, essentially a proxy score for how much environmental damage they're doing (and don't give me "solar" blah blah "solar" blah blah because the preponderance of them are anything but carbon neutral.
It's the tech equivalent to rolling coal.
Someone noted that it's easier- in theory- to upgrade or replace the equipment in an AI data centre, but the amount of power available is likely to be fixed once a data centre is built. Well, maybe.
I've posted most of this before, but I think the gigawatt-based metric is also symptomatic of something else. Something that even its opponents have give bizarrely little attention despite it being easily the biggest and most fundamental difference between the LLM AI boom and previous computer-driven revolutions.
It's the fact that the exponential increase in AI capacity in recent years has been predominantly driven by simply throwing similarly-increasing amounts of money and hardware at the problem- until we reached the current eyewatering levels- rather than the far more modest improvements in the underlying technology. In other words, brute force.
This is fundamentally different to what made the *original* computer revolution- from the end of WW2 to the present day- possible. That *was* driven almost entirely by such improvements in the underlying technology- those that Moore's law described. Exponential decreases in costs and scale of manufacture, coupled with the exponential increases in computing power they enabled.
There's no way this would have happened simply by building increasing numbers and bigger versions of those early valve/tube-based computer designs. There might have been a limited computer revolution of sorts (maybe with several computers in primitive "data centre"-like buildings per town), but you can only scale that "brute force" approach- and valve manufacturing- up so far before the cost and practicalities become prohibitive.
And the problem is that primitive "brute force" approach *is* the one that the current AI companies are relying upon.
Which is an issue because we're already clearly near the limits- the current bubble is *already* consuming ludicrous amounts of investment and it's just not likely to go *that* much higher.
And, to get back to the original point, the use of "gigawatt" as a crude synonym for computing power is a giveaway of this approach. Much like it would be in the "alternate future" described above, where the power of your local valve/tube-based computer was broadly proportionate to the amount of electricity it gobbled up.
“The Port Washington datacenter is being developed responsibly in partnership with the community, creating thousands of jobs, strengthening local businesses, and driving long-term economic growth across Wisconsin,"
Is there a single word in that statement that isn’t total bollocks?
As an aside, I love the phrase “negative free cash flow” as a euphemism for “losing money” :-)
That's to differentiate from losing money on an accounting basis, like if you write down the value of something and show a huge loss but you may still be making plenty of money on a cash basis and are easily able to pay your bills.
Negative free cash flow is something you can only sustain for so long, because you either need to have a lot of cash to start with or have creditors willing to continue to loan you cash to pay your ongoing obligations like salaries and debt service on your existing loans.
Poor cash flow breaks companies far faster than anything else.
A company can easily be asset-rich but go bankrupt because they don't have the free cash available to pay their employees, the IRS and suppliers. Some suppliers will accept an IOU for a while, but the rest won't.
Larry may need to choose between a yacht and a company rather soon. I suspect he'll go with yacht.
Agreed.
Especially the 'creating thousands of jobs' bit.
Even a 1GW datacentre campus is unlikely to have more than a couple of hundred permanent employees.
Although I imagine some will be quite well paid, given they'll be skilled trades like HVAC engineers, electrical engineers and so on.
Less so for security, FM and the guys that cycle round swapping out hard drives.
Exactly. I've joked- with only slight exaggeration- that data centres require two local employees, a security guard and his dog. The dog is there to bite the guard if he tries to touch any of the servers. (*)
As I, and countless other people keep pointing out, data centres might look like Big and Important investments in an area if you don't know any better, but in reality they're generally a shitty deal for communities. Even where they support high-skilled jobs, the vast majority of those will be using the facilities remotely with many based hundreds of miles or more away.
The actual number of skilled jobs- or jobs, full stop- created in the local area is likely to be limited relative to the massive disruption and demands placed on the local infrastructure, in addition to their being massive pig-fugly monstrosities.
(*) Yes, I ripped this off of a better joke about aeroplanes.
Just as is the problem with outsourcing in government and business, AI compute outsourcing must cost the newbies who are burning investor funds at an eye watering rate more than hosting their own processing faciliies. Of course they only need to convince greedy executives that they are as good as their word.
In three years time how many of the cloud and hosting companies will exist? They are the unregulated finance houses propping up over funded, poorly managed AI companies with the promise of AI profits being spent and ultimately the finance repaid over x years. If customers can be persuaded that there is trust and value from AI.
The markets are heading towards double or treble exposure as they pump billions into AI startups and in turn to the hosting companies those same AI companies are using as unsecured lines of credit.
AI startups are clearly expecting a notable gap between seed investment, IPO funding and customer revenues to such a degree that they are asking the US government to underwrite debt. With the greedy chump in the White House who knows where that will end.
What I feel could well happen is that a lot of institutions, individuals and sovereign states could find they are all lending to the same businesses with both on and off balance sheet debt and equity ratios the likes of which might never even have appeared in fictional works.
The fallout will surely make the dot com, banking and other global crashes seem like loosing a months pockect money.
