Google Says Its New Quantum Chip Indicates That Multiple Universes Exist (techcrunch.com) 56
Tucked away in a blog post about Google's quantum computing chip, Willow, Google Quantum AI founder Hartmut Neven wrote that the chip was so "mind-boggling" fast that it seemed to borrow computational power from other universes. According to Neven, the chip's performance suggests the existence of parallel universes, writing, "We live in a multiverse." TechCrunch reports: Here's the passage: "Willow's performance on this benchmark is astonishing: It performed a computation in under five minutes that would take one of today's fastest supercomputers 1025 or 10 septillion years. If you want to write it out, it's 10,000,000,000,000,000,000,000,000 years. This mind-boggling number exceeds known timescales in physics and vastly exceeds the age of the universe. It lends credence to the notion that quantum computation occurs in many parallel universes, in line with the idea that we live in a multiverse, a prediction first made by David Deutsch."
This drop-the-mic moment on the nature of reality was met with skepticism by some, but, surprisingly, others on the internet who profess to understand these things argued that Nevan's conclusions were more than plausible. The multiverse, while stuff of science fiction, is also an area of serious study by the founders of quantum physics. The skeptics, however, point out that the performance claims are based on the benchmark that Google itself created some years ago to measure quantum performance. That alone doesn't prove that parallel versions of you aren't running around in other universes -- just where the underlying measuring stick came from.
This drop-the-mic moment on the nature of reality was met with skepticism by some, but, surprisingly, others on the internet who profess to understand these things argued that Nevan's conclusions were more than plausible. The multiverse, while stuff of science fiction, is also an area of serious study by the founders of quantum physics. The skeptics, however, point out that the performance claims are based on the benchmark that Google itself created some years ago to measure quantum performance. That alone doesn't prove that parallel versions of you aren't running around in other universes -- just where the underlying measuring stick came from.
Uh oh (Score:3)
The multiverse might want it's computational power back.
Re: Uh oh (Score:5, Funny)
The apostropheverse wants its superfluous apostrophe back. It's means "it is".
Re: (Score:2)
Sorry, that was autocomplete.
Re: (Score:2)
Re: (Score:2)
Hmmm. The citations say that a Google spokesperson, observing a Google benchmark through a Google processor was so incredible that many zeros were needed to demonstrate the result.
Let's see if a third party, using a demonstrably repetitive and reliable rubric, is suddenly aided by a multiverse output.
It will be a SHTF moment for Google, deflating many, many zeroes.
Re: (Score:2)
Re: (Score:2)
Or at least to stop mining Bitcoin on their PCs.
Re: (Score:3)
Or at least to stop mining Bitcoin on their PCs.
A processor approaching this power should be able to break Bitcoin, putting a sudden end to this pestilence. As a bonus, most of the world's bad guys will be out huge sums of money.
Re: (Score:2)
I don't undestand how a grown adult can use the phrase "bad guys" earnestly. Cartoons have "bad guys." Reality has nuance.
Re: (Score:2)
I don't understand how you dont get the concept that "bad guys" are simply people that do bad things. It's a subjective term of course but for instance most would call a group like ISIS "bad guys" and that would make a lot of sense if you happen to be of the opinion that killing people for not believing in your god, being gay or for any number of other reasons is bad.
Re: (Score:2)
I don't get how, from my original complain, you think I don't undertand what 'bad guys" means. Of course I do. It's a term for little children.
My point was that it's a childish, borderline useless, term. Whenever I hear/see an adult use it it rings as "ethical assessment outsource to 3rd party."
Anyway. How could you possibly think I didn't understand the definition of "bad guy" ???
Re: (Score:3)
It's a moral generalization in a single term. I'm sorry it triggers you but there's nothing wrong with it.
Anyway. How could you possibly think I didn't understand the definition of "bad guy" ???
Because I thought you might not be being ridiculous and was giving you the benefit of the doubt.
Re: Uh oh (Score:2)
Re: (Score:2)
I imagine that the conductors of the simulation wonder what novel uses these little sentient agents are creating with their photons and electrons.
Also, the multiverse is just an side-effect of the parallelized nature of said simulation (they have to re-use the same particle in different cases).
Re: Uh oh (Score:2)
The F are you smoking? (Score:4, Insightful)
Where have I heard that one before?
Re: (Score:2)
"X did something faster than before, and is within expectations of X's capabilities, so Y must have been involved and therefore exists!"
Where have I heard that one before?
Maybe, but there is support from many in the scientific community for their findings. If it was a complete crackpot idea no one would be supporting further testing.
Re: (Score:2)
Maybe, but there is support from many in the scientific community for their findings. If it was a complete crackpot idea no one would be supporting further testing.
Have you paid any attention at all to the world in 2024?
Re: (Score:2)
My thought precisely. I'm with Sabine Hossenfelder in saying academia and science are close to jumping off a cliff and good riddance to bad rubbish.
We need a renaissance in this area because right now I'm at a point of distrust where I'd want to shake my other-dimensionly doppelganger's hand before I believe anything about multiverses.
Re: (Score:1)
How much do you want to bet that an AI pre-incarnation of Sabiney would be a rabid epicyclist?
This would take forever with a slide rule! (Score:4, Funny)
Some people at Google are hallucinating hard now (Score:2)
First, that "benchmark" does not say what they claim it says. Second, even if it did, that conclusion about parallel universes is complete bullshit and unfounded.
Ummmm (Score:4, Insightful)
No. Isn't this the same company with the dev who thought his AI was alive and conscious?
Re: (Score:2)
Re: (Score:2)
I have an engineering mentor who I used to work for. He's a brilliant coder and has rock solid engineering principles. But he's VERY into astrology. I could never square that circle about him.
Re: (Score:2)
Re: (Score:2)
I'm more of a "understand the core laws of the universe and be open to anything that fits within those laws. I also don't subscribe to the "if we don't know then it's possible" way of thinking. So with things like astrology I usually dismiss them outright. But I do agree that the human mind is tailor made to invent things and belief in the impossible is fundamental to human thought.
Re: (Score:2)
Re: Ummmm (Score:2)
Re: (Score:2)
No, not in the slightest.
Re: Ummmm (Score:2)
Re: (Score:2)
You can't be this naive. LLMs aren't some tech that just sprang out of nowhere. It has been developed over a course of decades. There are dozens of publicly available research papers out there to read. Anyone with half a brain can understand the basics of how it works and understand that it is not intelligence. Only someone delusional would think otherwise. So again, no, I would never confuse it for anything other than a glorified lookup table with a mind-bogglingly large indexed dataset.
Re: Ummmm (Score:2)
Re: (Score:2)
So we can look forward to Hartmut Neven being fired?
Great Scott ! (Score:2)
To run this miracle of a processor, you're gonna need something with a little more kick to generate the 1.21 gigawatts of electricity it needs. . . . . :|
. . . . and probably an entire lake just to keep it cooled
Hugh Everett? (Score:1)
in line with the idea that we live in a multiverse, a prediction first made by David Deutsch
..was it not Hugh Everett that first proposed a many-worlds/multiverse interpretation, rather than David Deutsch? (I know super little about this and am probably showing my ignorance - David Deutschs work seems more directly relevant to computation?)
https://en.wikipedia.org/wiki/Hugh_Everett_III/ [wikipedia.org]
Kind of a problem (Score:1)
It lends credence to the notion that quantum computation occurs in many parallel universes
Hopefully not true, as if so it means you have basically built a terminal for a timeshare system.
Which means as other worlds in the multiverse also build the same quantum computation, it will slow down our quantum computers - possibly even causing contention over shared resources
You can imagine that as some point a galactic IT worker is going to come through a portal and demand we pay up for stolen resources.
Re: (Score:2)
It lends credence to the notion that quantum computation occurs in many parallel universes
Hopefully not true, as if so it means you have basically built a terminal for a timeshare system.
Which means as other worlds in the multiverse also build the same quantum computation, it will slow down our quantum computers - possibly even causing contention over shared resources
I guess we can look forward to an infinite number of infinity wars! Maybe we're already in one and just don't realize it yet.
Re: (Score:2)
I have to say (Score:2)
This has to be one of the most asinine statements I've read in a long time. Was this story actually an April 1 holdover?
But honestly, I'm guessing this was less a serious statement and more a case of rah-rah marketing blathering than anything else - given the statement was made by the co-founder of the group involved.
Re: (Score:1)
Does the strong emotional content of your post say more about you than about multiverses?
Technobabble (Score:2)
All I want to know is if this thing can be harnessed to get me into the universe in which Star Trek was not ruined!
A More Likely Conclusion (Score:4, Interesting)
Posts like this are not signs of new physics. They are signs that the underlying technology is fruitless and functionally a scam. It's also a sign that people running the programs have become shut-eye [youtube.com] and believe their own hype. Much like fusion power, AI, self-driving cars, biofuels, clean coal, hydrogen fuel cells, carbon-capture, and quantum computing isn't real, it's the sincere hope of the rich and powerful who want to keep being rich and powerful.
If something can't go on forever, it won't. There are real, physical limits to what you can do in a society determined by how much energy that society can capture. Fusion power lets you get around those limits without changing anything else about how society functions. It would be nice for those who enjoy the status quo but it just isn't so. Fusion power was, and will forever remain, twenty years in the future.
There are real, physical limits on the number of cars a society can accommodate. Not least of which is the limit of people who can both afford to purchase and maintain a vehicle and are also capable of legally operating it. When you hit that limit, car sales will flatten or shrink. Self-driving cars let you get around those limits and thus capture an entirely virgin market. That's great if you are a car manufacturer (or if you're in the financial sector around it that is actually several times its size). If it was going to happen, it would have happened anytime in the last two decades of it being a few years away.
Fossil fuels are great. Lots of energy stored in a compact and convenient package that stores well and is easy to transport. However, the consequences of its use are dire. Biofuels, hydrogen fuel cells, and clean coal let you keep that infrastructure and those supply chains largely unchanged while avoiding or delaying those consequences. That would be great if you are in the most profitable industry in human history. The fact that the numbers don't add up won't stop those people from wishing they did.
The computer industry has been fabulously successful. It's made a lot of people very wealthy. One of the reasons for this is that Moore's Law means nearly everyone who uses a computer functionally has to replace that computer every five years. If you are a power user, you probably have to replace your computer every two years at a premium price. But Moore's Law isn't a physical law; it's an economic law. It's always been about how often can you reasonably make a person buy a new computer before opportunity costs dictate they stick with what's working for now and wait a little bit to get a much better option. The hard, physical limits of the problem of increasing computational density have slowed the ability of the industry to keep pace with that economic reality. If the average user goes from buying a new computer every five years to buying a new computer every five and a half years, the market contraction that would represent would devastate the entire industry. But if quantum computers come out, everyone needs to buy all new computers right now. Even if you don't use quantum processes, you'd need new classical computers designed with the knowledge that quantum computers are out there and have to be accounted for in the security of your system. That would be great for the industry. To be clear, it hasn't been shown to actually work. Certain parts of it have somewhat worked in isolation. The more dramatic promises are, as yet, vaporware.
Bullshit (Score:2)
There is nothing in this article that even remotely pertains to any sort of proof that backs up the claims. This story is about as valid as that one time a Google researcher said that their chatbot was sentient. Utter bullshit.
https://www.scientificamerican... [scientificamerican.com]
Ether that ... (Score:2)
OR the wavefunction collapsed completely in line with the Copenhagen interpretation. Excellent! We've now got solid proof that our various interpretations of quantum physics are indeed varied and interpretations. Good work, gentlemen.
So... (Score:2)
Nobel prize (Score:2)