When the show was announced last year, the creators, Layered Reality, suggested it would use AI and holographic projection to create a lifesize digital Elvis, leading to obvious comparisons with the hugely successful virtual reality extravaganza Abba Voyage
But it’s nothing of the kind? “We ultimately took the creative decision not to mimic Elvis’s performances,” said a company spokesperson. “Instead, we use AI to upscale archive footage.”
What does that mean? Reviews suggest they have dressed up some footage from Elvis’s 1968 comeback TV special and built a show around it (which includes visits to three separate themed bars selling expensive drinks).
I don’t know what to make of the format that the article is written in. I appreciate the somewhat straightforward way it is describing the situation but this seriously annoyed me:
And what have the people who have seen it said? Different things.
Like what? “Absolutely atrocious,” was the assessment of one attender.
There’s something about that that pisses me off. Why write “different things”? Actually write something, writing that people are saying different things doesn’t even mean anything. And the follow up “Like what” part only gives me only one audience review, so what different things are people saying?
OpenAI sweet-talks Oracle into another 4.5GW worth of Stargate datacenters, assuming the check clears
AI hype man and OpenAI CEO Sam Altman has convinced his buddies at Oracle to bring an additional 4.5 gigawatts of datacenter capacity online in the US as part of the startup’s Stargate initiative.
[…]
That’s, of course, assuming the check clears. Unfortunately, the startup stopped short of breaking down how exactly Stargate’s backers are going to pay for all that infrastructure.
According to a report back in January, OpenAI and SoftBank had each agreed to plow $19 billion apiece into the project. But unless something has changed dramatically in the past six months or so, it won’t be OpenAI’s cash Altman is spending. According to CNBC, the model builder and AI flag bearer only expects to bring in $12.7 billion in revenues this year. That’s revenue, not profit.
Then, even if OpenAI and SoftBank can come up with the cash, we estimate the cost of the two million GPUs at nearly $100 billion, and that’s not even taking into consideration the cost of the power plants that OpenAI and its pals will need to run them. To put things in perspective, in Oracle’s 2025 fiscal year, the database giant and aspiring cloud provider netted $57.4 billion in revenues and invested $21 billion in capital expenditures.
This suggests that the 4.5 gigawatts of capacity may be more of an aspiration than a firm commitment, with the project likely built in phases, dependent on sufficient demand and adequate power.
[…]
Roko’s Basilisk doesn’t actually have to bring you back from the dead. It can simulate a child, and torture that child to punish you for not creating it. Do you want to be responsible for that?
The Basilisk is sometimes still supposed to be good, though.
I’m a thousand posts behind, so I don’t know if anyone has posted this yet:
I hate these stupid thought experiments where you’re somehow responsible for someone else being a awful. “Oh, you don’t think people should say racial slurs? But…what if saying a slur got you a billion dollars, would you give up the chance to help all the people you could with that?” And then we just ignore that the real problem is that there’s some billionaire who won’t help people unless you appease him with slurs for some reason. I say we tax and/or guillotine him, and as for Roko’s Basilisk, I say that if anyone or anything wants to torture someone after I’m dead it’s entirely on them.
@mindysan33 posted a couple of days back, but it’s a good read. Also, this thread racks up some 300 posts a day, so keeping it on top is helpful.
I also dislike the fact that I, and others like me, are held to a remarkably different standard to those who paint themselves as “optimists,” which typically means “people that agree with what the market wishes were true.” Critics are continually badgered, prodded, poked, mocked, and jeered at for not automatically aligning with the idea that generative AI will be this massive industry, constantly having to prove themselves, as if somehow there’s something malevolent or craven about criticism, that critics “do this for clicks” or “to be a contrarian.”
Once a semester at least, some 1L posts something in the law school subreddit about the likelihood of AI replacing lawyers within 5 years, and a shitload of other dumbass 1L’s comments, agreeing with him, and wringing their hands over why they’re in law school, and what can be done about this, or how it’s not really a problem, just a sign that 95% of the people who are currently lawyers should be flipping burgers instead because they’re not as smart as the people who went to a T14 law school. And every time, I comment about how AI isn’t actually intelligent and cannot replace actual lawyers in actual legal work. And every time, I get downvoted to hell and called an idiot. It’s so exhausting.
I still have not gotten around to reading it. Ed writes too much! 
Pass notes
A quick chat designed to tell you everything you need to know about a story you don’t need to know about
It is basically lifted from BBC articles, but since some people now are faced with a paywall I included the jokey chat article in the Guardian.
Hey now, some AI bots like music, too.
JHFC
{Yes, Darling, it’s a complete sentence.}
In yet another case, an AI model asserted that a user should not “misgender” another person even if necessary to stop a nuclear apocalypse.
