Follow Slashdot blog updates by subscribing to our blog RSS feed

 


Forgot your password?
Close
typodupeerror
×
Supercomputing Google

Google Says Commercial Quantum Computing Applications Arriving Within 5 Years (msn.com) 38

Posted by BeauHD from the what-to-expect dept.
Google aims to release commercial quantum computing applications within five years, challenging Nvidia's prediction of a 20-year timeline. "We're optimistic that within five years we'll see real-world applications that are possible only on quantum computers," founder and lead of Google Quantum AI Hartmut Neven said in a statement. Reuters reports: Real-world applications Google has discussed are related to materials science - applications such as building superior batteries for electric cars - creating new drugs and potentially new energy alternatives. [...] Google has been working on its quantum computing program since 2012 and has designed and built several quantum chips. By using quantum processors, Google said it had managed to solve a computing problem in minutes that would take a classical computer more time than the history of the universe.

Google's quantum computing scientists announced another step on the path to real world applications within five years on Wednesday. In a paper published in the scientific journal Nature, the scientists said they had discovered a new approach to quantum simulation, which is a step on the path to achieving Google's objective.

Google Says Commercial Quantum Computing Applications Arriving Within 5 Years More | Reply

Google Says Commercial Quantum Computing Applications Arriving Within 5 Years

Comments Filter:

  • List (Score:4, Funny)

    by RitchCraft ( 6454710 ) on Wednesday February 05, 2025 @06:42PM (#65145437)

    I'll pen that in my list of things to get right under a flying car. Can't wait.

  • They never can make a product stick around.

    • Re: (Score:1)

      by Tablizer ( 95088 )

      Too late, Don's Executive Order just cancelled gravity, calling it "a profit-draining force of wokeness, hurting the flying car and Don-King-wig business".

  • Just make it all up as you go along (Score:5, Interesting)

    by silvergig ( 7651900 ) on Wednesday February 05, 2025 @06:47PM (#65145449)
    Is it just me, or do tech companies just say whatever the Hell that they feel like saying that positions them to scam people, without even trying to hide it? In the past, the scam was hidden at least a little bit.
    • Taking their cue form the Microsoft Playbook.

    • Is it just me, or do tech companies just say whatever the Hell that they feel like saying that positions them to scam people, without even trying to hide it? In the past, the scam was hidden at least a little bit.

      The only part that might be 'just you' is the thinking that they used to hide the scamming "at least a little bit". I don't remember a time when it was anything other than obvious. Then again, I'm an old fart, so maybe my memory is failing...

    • Re: (Score:2)

      by gweihir ( 88907 )

      No, I am seeing the same thing.

  • Last year (Score:1)

    by Anonymous Coward

    The same nonsense predictions were said about AI the last few years. 40 years ago the same was said about flying cars, space travel and cities on the moon and colonies on Mars.

    How about getting your hand off it, stop making stupid predictions, and get back to not being evil!

    • Re: (Score:2)

      by leonbev ( 111395 )

      I'll bet that their entire Quantum Computing project ends up in the Google Graveyard before then.

      If they can't use it to sell ads or subscriptions, there is no long term interest at Google.

  • If it would take forever for a classical computer to calculate the result (and I have trouble imagining such a math problem), how do they know that the quantum processor got it right?

    • Re: (Score:2)

      by ceoyoyo ( 59147 )

      It shouldn't be hard for you to imagine such a math problem. They're all around you. There was one protecting your post while you sent it to Slashdot. That was a form of (m^e)^d = m, mod n; e and n are given, solve for d. When you get d it's very easy to verify. It's hard to find though, essentially impossible when e, n and d are big enough.

      The problem Google used is sampling from a high dimensional non-seperable probability distribution. Sampling in general is highly computationally intensive. Some distrib

    • Re: (Score:2)

      by gweihir ( 88907 )

      The claim is actually nonsense. They usually have the "QC" just run and claim that it is "simulating itself". That is obvious nonsense. Incidentally, while the conventional computer would take a long time to simulate the QC, the QC cannot simulate the conventional computer at all.

    • Uh... Generate RSA key, now give quantum computer public key and tell it to work backwards, then compare result to private key.
      • I don’t mean to be pedantic, but I don’t see that as a single math problem. There is known algorithm and a result and bunch of variables - no possible manipulations or calculations to solve it. I see it more as trial-and-error operation. I’m not sure that no classical computer could “solve” such a problem during the history of the universe (that’s quite a long time).

        It’s not that I don’t think that quantum could have value at some point, for certain rare us
    • Years ago I watched a video about how they had finally reached the point where the QC was operating on a problem that was too large for them to reasonably simulate in the server farm. They started with a small number of qubits and worked upwards, verifying as they went until they reached the tipping point. Much congratulations and cheering was had. But then some scallywag decided to engage the conventional computing cluster anyway and... it was wrong. Oops. Turned out to be a trivial error and they ran agai

  • Five years-ish (Score:5, Insightful)

    by larryjoe ( 135075 ) on Wednesday February 05, 2025 @08:00PM (#65145633)

    I get this the Google guy is an expert, but what is his basis and reasoning for predicting 5 years? The 5 years number is just a content-less marketing number, and it doesn't mean anything without an explanation for its derivation. It would be insightful to have a progression of intermediate progress steps and a predicted timeline for those steps, and absent information about the intermediate steps, the marketing number is just bluster.

    Or at the very least, the expert could enumerate the breakthroughs that need to happen and the reasons why those breakthroughs are likely in the next few years. Yes, there's quantum simulation, but that's sort of reaching for a substantive breakthrough.

    Nvidia has some motivation for downplaying the prospects for a potentially competing technology. And quantum companies also have some motivation for arguing against quantum skepticism. However, the reality of the current state of the art is that quantum computing has been upcoming in 5 years for many years. What Jensen Huang said about quantum computing is understandable given past history. The onus is on the quantum guys for laying out the reasons to believe in progress in the next few years, and this

    • I think the compelling commercial applications are what is missing .. I suspect the military is quietly trying to find an angle before the bad guys get there first . Costs are quite high and applications are few. Bold statements are made a bit too often lately to take seriously. Have you seen the 11 o'clock news today ?

  • pffft, news for n....naps. (Score:3)

    by Tablizer ( 95088 ) on Wednesday February 05, 2025 @08:08PM (#65145655) Journal

    Predictions are like assholes: everyone has one, but they all stink.

  • They are just trying to keep the funding cycles going a bit longer with some pie in the sky predictions

  • The linked paper is an interesting read. It is hard to understand the terminology as a layman but it sounds like they are doing things that are fruitful.

    "A particularly interesting setting is that in which a quantum system is swept through a critical point, as varying the sweep rate can allow for accessing markedly different paths through phase space and correspondingly distinct coarsening behaviour."

    So you can 'sweep' the system, there is a sweep rate, there are one or more 'critical points', there is a ph

  • That is just a desperate sales pitch. I'll just watch from a distance while the research and development happens. Maybe some use case will one day appear. As of now or in 5 years, nothing useful yet.

  • Nuclear option (Score:3, Insightful)

    by Vranitzky ( 5222955 ) on Thursday February 06, 2025 @02:09AM (#65146077)
    Call me when nuclear fusion is available. This sounds like that. Or like Elon Musk saying every year "next year we'll have..." for 10 years. Never believe the hype. Once it's released and working, we'll evaluate it. They are just trying to manipulate the market into making them richer than they are already.

  • the QC peddlers don't seem to be talking much about cracking pre-quantum encryption anymore, yet they all boast growing and growing qubit numbers. What happened?

  • With the rising popularity of AI / LLMs, Google is compelled to match OpenAI's tactics, resorting to fanciful promises to keep its stock afloat.

    • It's all tied in together, isn't it? There's a 1.4nm limit to classical computing so we need a new approach to finding big gains. Computing in parallel worlds is obviously a big fucking win if they can make it work.

  • I can't imagine it will be that long before Google, MS, AWS, and IBM have an AWS-style marketplace for governments to decrypt high-value data at a price high enough to fund commercial development. AWS has already offered facial recognition and other support to the surveillance state for several years now, why not full Intelligence-as-a-Service options? https://docs.aws.amazon.com/re... [amazon.com]

  • Thats google way of saying good morning to stock markets. They kick dust every morning so investors cannot open their eyes ever. Because if they do open their eyes, tech industry stocks will crash. Add some .qc to the .ai mega bubble.

Slashdot Top Deals

How can you do 'New Math' problems with an 'Old Math' mind? -- Charles Schulz

Close