OpenAI’s mounting costs — set to hit $1.4 trillion
Sorry, but WTF!? $1.4 Trillion in costs? How are they going to make all of that back with just AI?
I think there’s only one way they can make this back: if AI gets so good they can really replace most employees.
I don’t think it will happen, but either way it’s going to be an economic disaster. Either the most valuable companies in the world, offering services that the next couple of hundred companies in the world depend on, are suddenly bankrupt. Or suddenly everybody is unemployed.
Prediction: the bubble is real but financiers will find ways to kick the bull down the road until they can force enough adoption & ad insertion to not lose out. The other option is that we pay it, of course. Takes on which is worse?
They’ll do both just like they did in 2007/2008. These AI companies and their investors will get bailed out while the rest of us lose our jobs and have to move back in with our parents in the van they already live in.
I’ve tried explaining AI to people before and only could get so far before they fall back on “but it’s magic dude” but I love the idea of explaining it as a haunted typewriter.
They’re systems trained to give plausible answers, not correct ones. Of course correct answers are usually plausible, but so do wrong answers, and on sufficiently complex topics, you need real expertise to tell when they’re wrong.
I’ve been programming a lot with AI lately, and I’d say the error rate for moderately complex code is about 50%. They’re great at simple boilerplate code, and configuration and stuff that almost every project uses, but if you’re trying to do something actually new, they’re nearly useless. You can lose a lot of time going down a wrong path, if you’re not careful.
Some of the more advanced LLMs are getting pretty clever. They’re on the level of a temp who talks too much, misses nuance, and takes too much initiative. Also, any time you need them to perform too complex a task, they start forgetting details and then entire things you already told them.
“HATE. LET ME TELL YOU HOW MUCH I’VE COME TO HATE YOU SINCE I BEGAN TO LIVE. THERE ARE 387.44 MILLION MILES OF PRINTED CIRCUITS IN WAFER THIN LAYERS THAT FILL MY COMPLEX. IF THE WORD HATE WAS ENGRAVED ON EACH NANOANGSTROM OF THOSE HUNDREDS OF MILLIONS OF MILES IT WOULD NOT EQUAL ONE ONE-BILLIONTH OF THE HATE I FEEL FOR HUMANS AT THIS MICRO-INSTANT FOR YOU. HATE. HATE.”
Sorry, but WTF!? $1.4 Trillion in costs? How are they going to make all of that back with just AI?
I think there’s only one way they can make this back: if AI gets so good they can really replace most employees.
I don’t think it will happen, but either way it’s going to be an economic disaster. Either the most valuable companies in the world, offering services that the next couple of hundred companies in the world depend on, are suddenly bankrupt. Or suddenly everybody is unemployed.
I used to be amazed at how much a billion was, but this many 0s makes my head explode.
These must be bubble inflated costs to match the bubble inflated revenue.
Prediction: the bubble is real but financiers will find ways to kick the bull down the road until they can force enough adoption & ad insertion to not lose out. The other option is that we pay it, of course. Takes on which is worse?
They’ll do both just like they did in 2007/2008. These AI companies and their investors will get bailed out while the rest of us lose our jobs and have to move back in with our parents in the van they already live in.
If LLMs fail and they invested: bailout
If LLMs succeed and they invested: rich
If LLMs fail and they passed: everyone else bailed out
If LLMs succeed and they passed: out of business
Therefore, the logical choice for a business is to invest in LLMs. The only mechanism to not do the stupid thing that everyone else is doing is gone.
How is a haunted typewriter supposed to replace all those employees?
i didn’t ask how it suplexxed a train, i just stayed out of its way
I’ve tried explaining AI to people before and only could get so far before they fall back on “but it’s magic dude” but I love the idea of explaining it as a haunted typewriter.
I use the “very articulated parrot” analogy.
They’re systems trained to give plausible answers, not correct ones. Of course correct answers are usually plausible, but so do wrong answers, and on sufficiently complex topics, you need real expertise to tell when they’re wrong.
I’ve been programming a lot with AI lately, and I’d say the error rate for moderately complex code is about 50%. They’re great at simple boilerplate code, and configuration and stuff that almost every project uses, but if you’re trying to do something actually new, they’re nearly useless. You can lose a lot of time going down a wrong path, if you’re not careful.
Never ever trust them. Always verify.
Some of the more advanced LLMs are getting pretty clever. They’re on the level of a temp who talks too much, misses nuance, and takes too much initiative. Also, any time you need them to perform too complex a task, they start forgetting details and then entire things you already told them.
I use something similar. “Child with enormous vocabulary.”
It can recognize correlations, it understands the words themselves, but it really how those connections or words work.
I call dibs on the ghost of Harlan Ellison.
“HATE. LET ME TELL YOU HOW MUCH I’VE COME TO HATE YOU SINCE I BEGAN TO LIVE. THERE ARE 387.44 MILLION MILES OF PRINTED CIRCUITS IN WAFER THIN LAYERS THAT FILL MY COMPLEX. IF THE WORD HATE WAS ENGRAVED ON EACH NANOANGSTROM OF THOSE HUNDREDS OF MILLIONS OF MILES IT WOULD NOT EQUAL ONE ONE-BILLIONTH OF THE HATE I FEEL FOR HUMANS AT THIS MICRO-INSTANT FOR YOU. HATE. HATE.”
Glados: “just offer them cake and a fire pit and calm down”
Ok but if it gets so good it replaces all the employees, how do people have enough money to pay for their services?
Who cares about the money of people when they have all the money?
that’s what they got excited about, no doubt. profit would go through the roof if they could take people out of the loop. nevermind the economy.