My cousin sent me this today, and after reading the article I've got some mixed feelings. On one hand, quantum computing is an incredible technology that seems like it can push us further forward....
Today I’m delighted to announce Willow, our latest quantum chip. Willow has state-of-the-art performance across a number of metrics, enabling two major achievements.
The first is that Willow can reduce errors exponentially as we scale up using more qubits. This cracks a key challenge in quantum error correction that the field has pursued for almost 30 years.
Second, Willow performed a standard benchmark computation in under five minutes that would take one of today’s fastest supercomputers 10 septillion (that is, 1025) years — a number that vastly exceeds the age of the Universe.
My cousin sent me this today, and after reading the article I've got some mixed feelings.
On one hand, quantum computing is an incredible technology that seems like it can push us further forward. The blog casually mentions parallel universes, which is kind of wild to see from such a major company like Google. Quantum physics has always been an exciting concept for me and having it become 'real' and tangible is fascinating.
On the other hand, I feel that humanity is already struggling to keep up. I feel a sentiment of "just because we can, do that mean we should?" It occurs to me that a 256-bit quantum computer could likely break a lot of modern encryption algorithms, and trying to get people to understand the quantum paradigm when most already struggle to grapple with classical computation could further alienate people from modern technology.
These are well-considered thoughts! I respect the depth you put into this. I don't worry too much about codebreaking, for a bunch of reasons. Most serious symmetric cryptography (so, government...
These are well-considered thoughts! I respect the depth you put into this.
I don't worry too much about codebreaking, for a bunch of reasons. Most serious symmetric cryptography (so, government stuff) is probably resilient to quantum; organizations are already implementing post-quantum algorithms; and generally speaking, breaking encryption in one place doesn't get you all that much - you need to be able to do it all over, and even with a very fast quantum computer, that will take a long time.
What I've never considered before is your suggestion that people are already alienated from technology and quantum (which is an order of magnitude more complicated will make it worse). This is a really interesting thought. Ultimately, what I wonder is whether anybody really understands modern technology anyway. The best of us maybe have mastery over one field, but that field is probably pretty narrow. (How many excellent software engineers know anything about UV lithography? For that matter, how many excellent software engineers know anything about writing code for hardware?)
From that standpoint, I'm not sure quantum computing will make things too much worse - but I do think that there are likely to be quite a few interesting benefits.
To address the main point of interest, I'll qualify myself by saying that I have a degree in Computer Science and at best a vague understanding of anything that happens below compiler level. I...
To address the main point of interest, I'll qualify myself by saying that I have a degree in Computer Science and at best a vague understanding of anything that happens below compiler level. I totally understand that the world of tech (perhaps science as a whole) is built on layers of abstraction, and at the same time I think that the world of open-source is hugely comforting for me. I like to know that I am depending on as few "black boxes" as possible. I don't understand UV lithography, but I know that I could if I wanted to. I cannot say the same thing about Windows or Squarespace or Instagram, so my philosophy is to suggest that people steer clear of them.
Personally, I spend a lot of my time analyzing social media and digital identity on the internet, and a big part of what I aspire to do is teaching people how to "own their identity" in the digital world. I think that basic digital literacy is hugely lacking, and I already struggle to concisely explain in simplest terms what a website is and how it works.
As an 'ambassador of technology', people ask me what programming languages are or how domain names work, and I can usually answer those questions sufficiently. I've got a handle on explaining LLMs now. I don't really understand the implications of quantum computing, however, so I would be at a loss about how to explain this article to someone in layperson terms.
In general, I would agree that nobody understands everything about modern science. However, I believe that there is a certain degree of fluency that can be attained through study, and that sort of fluency is what enables people to pursue autonomy in the world. I believe that autonomy and ownership is what helps people confront the feelings of alienation, and we are already at a point where people do not feel a real sense of ownership over their online presence. This post isn't really mine, for example, as it doesn't live on my computer. I might make a copy of it, but probably not.
Seeing this article has made me seriously consider the household quantum chip, and how that will fit in to this larger image that I'm painting. What will people use it for? How will it be 'sold' to us? Will quantum technology become a cornerstone of the next scientific paradigm? Will we be able to leverage concepts such as quantum entanglement to share information over vast distances?
At the same time, will we even know what to do with it? The technology we already have in our hands could be used to create worldwide support networks and address problems on a global scale, but instead we use it to drive corporate profit and cope with the profound loneliness of our modern day by dissociating into constructed realities. If we bring these same habits and attitudes into the quantum world, I'm concerned that these issues will also be amplified by an order of magnitude - and I don't want to imagine what that might look like.
I think you’re crossing into science fiction. Quantum computing will allow us to do certain types of calculations very quickly. That’s it. Quantum entanglement cannot transmit information.
Will quantum technology become a cornerstone of the next scientific paradigm? Will we be able to leverage concepts such as quantum entanglement to share information over vast distances?
I think you’re crossing into science fiction. Quantum computing will allow us to do certain types of calculations very quickly. That’s it.
It can't transmit information faster than light, but quantum channels still seem like an active field of research! I think there's lots of stuff we have today that was once the realm of science...
It can't transmit information faster than light, but quantum channels still seem like an active field of research! I think there's lots of stuff we have today that was once the realm of science fiction.
Complete sidebar, but UV lithography is pretty neat! Asianometry has done some excellent explainer videos around the topic (bouncing lasers off of molten tin droplets! So cool :D), and these days,...
Complete sidebar, but UV lithography is pretty neat! Asianometry has done some excellent explainer videos around the topic (bouncing lasers off of molten tin droplets! So cool :D), and these days, old photolithography processes are within grasp of exceptionally interested hobbyists! You don’t even need to be the child of a wealthy Silicon Valley tech person to do so :3
The problem is not that symmetric encryption is quantum resistant it's that TLS key agreement today isn't and governments today are recording traffic with the intent to decode it later, which they...
The problem is not that symmetric encryption is quantum resistant it's that TLS key agreement today isn't and governments today are recording traffic with the intent to decode it later, which they will be able to do, by attacking key agreement and then using the recovered symmetric key on the rest of the stream. I guess I only hope my own traffic isn't interesting enough for anyone to bother with.
I'm not saying that this isn't happening in some limited instances, because it definitely is. But there are a ton of challenges associated with this. Collection - practical: collecting all traffic...
governments today are recording traffic with the internet to decode it later, which they will be able to do, by attacking key agreement
I'm not saying that this isn't happening in some limited instances, because it definitely is. But there are a ton of challenges associated with this.
Collection - practical: collecting all traffic worldwide ever is much harder than you think. You need to either secretly control key nodes or force other entities to do it for you. How would the NSA collect random internet traffic in Finland, for instance? Even if we think that they control every node worldwide (they don't), they'd then have to send copies of all this traffic back to their data centers, meaning that all internet traffic would be effectively doubled. I think someone would probably notice.
Collection - legal: I know this sounds maybe a little laughable, but most Western surveillance agencies are prohibited from collecting on their own citizens without a warrant. Again I'm not saying it never happens, but having to do this all secretly makes it an order of magnitude harder.
Storage: How could you possibly store this much data? Think of how much data is generated every day - petabytes and petabytes. You can't tell what most of it is (because it's all encrypted) so you need to store all of it. That's impossible.
Archiving and search: okay, let's say I'm wrong and the NSA is secretly storing all your data. How do they know it's yours? All they would have is 'public IP address 1 talks to public IP address 2,' and those change pretty frequently. It makes searching through this stuff absolutely impossible.
Fun bonus challenges: even encryption susceptible to quantum decryption often includes things like the double ratchet algorithm that forces you to decrypt every exchange one by one. So let's say your quantum computer can decrypt a WhatsApp message every microsecond. That's pretty good! ...but turns out messages are being generated ten times faster than you can decrypt them, so you just fall farther and farther behind. Rats.
Ultimately, you have something that's extremely difficult to collect, impossible to store in bulk, maddening to search through, ...and it's all illegal.
I'm not saying this never happens. But I am saying that there's absolutely no way that it's happening in bulk for regular people.
I never said anything about anyone collecting all the traffic in the world at once, and I explicitly called out that I think someone would have to find me interesting for this to be a problem for...
I never said anything about anyone collecting all the traffic in the world at once, and I explicitly called out that I think someone would have to find me interesting for this to be a problem for me.
I do know how hard this would be. I used to work for a company that produced physical network taps for ixp fibre cross-connects. Lots of companies are already engaged in interface mirroring for security purposes with a rounding error on government budgets. My current company (mundane IoT) has port mirroring on internet facing ports for troubleshooting because it's significantly more effective to just chuck packets into object storage and come back for them if we need to look at them than it is to log all the derived data and throw the traffic away.
Suggesting that it would be hard for large-scale dragnet domestic sigint programs to exist in secret for years post-Snowden is a bold take.
Most harvest now, decrypt later operations are probably targeted and already know what they are after so identifying the traffic has been done upfront.
(Minor comtext, but I'm not American, I'm in the UK)
Btw one last thing on double ratchets: the security of key rotation / forward secrecy is predicated on the idea that the key agreement steps cannot be attacked. But if they are attacked and you...
Btw one last thing on double ratchets: the security of key rotation / forward secrecy is predicated on the idea that the key agreement steps cannot be attacked. But if they are attacked and you continue to see the complete stream of messages then you can simply mirror the stream of operations either alice or bob is performing without having to perform the attack every time because you acquired one of the dh secret keys + seed material. Signal isn't magic.
Seems fine to me. Ultimately a quantum computer just runs specific algorithms at a different complexity class (ie “way, way faster”) than a classical computer can. One of those quantum algos is...
Seems fine to me. Ultimately a quantum computer just runs specific algorithms at a different complexity class (ie “way, way faster”) than a classical computer can. One of those quantum algos is capable of solving some of our popular encryption algorithms, but ultimately this is a known problem with solutions in progress.
Try not to do anything illegal and encrypt evidence of it in the next twenty years, but otherwise, this isn’t a huge deal. Twenty years ago most wifi connections were insecure and no one encrypted their network traffic. The goal behind traffic encryption has always been to make it seamless enough that people will actually use it, not to have them understand it mathematically, I feel, so this will be no different.
Re. The parallel universe and AI stuff, afaik it’s a combination of philosophical interpretations of the mathematics underlying quantum mechanics and hype chasing. This person is legally obligated to mention AI at least twice in all public communications.
Yeah, as a physicist I'll just say, no, this google chip does not create parallel universes in any meaningful way.
The parallel universe and AI stuff, afaik it’s a combination of philosophical interpretations of the mathematics underlying quantum mechanics and hype chasing.
Yeah, as a physicist I'll just say, no, this google chip does not create parallel universes in any meaningful way.
Cool, thanks for chiming in. As someone who doesn't really know his stuff, my mental model is that parallel universes are not created or destroyed, but something like concurrent processes which we...
Cool, thanks for chiming in. As someone who doesn't really know his stuff, my mental model is that parallel universes are not created or destroyed, but something like concurrent processes which we are constantly moving between based on observation. I imagine that quantum technology allows us to determine what the outcome of an observation will be in a small cross-section of higher dimensional space and select the one with a result that we're searching for, with more powerful quantum computers being able to cast a larger field. Is any of this provable, falsifiable, or addressed by current theories?
I’d note that MWI is increasingly controversial, if nothing else because it relies on the linearity of quantum mechanics and many attempts to reconcile gravity and quantum mechanics create...
I’d note that MWI is increasingly controversial, if nothing else because it relies on the linearity of quantum mechanics and many attempts to reconcile gravity and quantum mechanics create non-linear systems.
But even in MWI, you don’t move between parallel worlds. You simply are in one world. When there is a superposition described by probability, what is unknown is what world you are in, not the result, based on your current information.
I imagine that quantum technology allows us to determine what the outcome of an observation will be in a small cross-section of higher dimensional space and select the one with a result that we're searching for
No, it just lets you do types of math very quickly.
MWI is like fundamentally untestable since a basic underlying principle is that each world cannot interact with other worlds.
"Try not to do anything illegal" is a very concerning statement for me, mostly because we are living in an era where fascist ideologies are hot and trending. Twenty years ago, there was no...
"Try not to do anything illegal" is a very concerning statement for me, mostly because we are living in an era where fascist ideologies are hot and trending. Twenty years ago, there was no widespread attention economy and the internet was a much different place.
Anyways, I do actually agree with the Many Worlds hypothesis and I consider quantum technology to be something that could allow us to leverage higher dimensions than the ones our physical bodies occupy. What we would do with that, I still don't know! But I hope we can direct it towards making this world a better place for all the beings that inhabit it.
Certainly, and agreed! But that said, if we assume for a future fascist dictatorship, evidence of misgivings will be unnecessary for arbitrary punishment. “Try not to record yourself breaking the...
"Try not to do anything illegal" is a very concerning statement for me, mostly because we are living in an era where fascist ideologies are hot and trending. Twenty years ago, there was no widespread attention economy and the internet was a much different place.
Certainly, and agreed! But that said, if we assume for a future fascist dictatorship, evidence of misgivings will be unnecessary for arbitrary punishment. “Try not to record yourself breaking the law” is already a tough bar for many folks to clear (lighter side: people doing ‘social experiments’ on YouTube/TikTok, darker side: existing as LGBTQ+ and being doxxed), so I doubt that breaking encryption will be directly importantly for the overwhelming majority of us. Maybe it’ll matter more for three letter agencies.
[many worlds interpretation]
I’m more of a pilot wave theory person myself, but I can see the appeal behind the other interpretations! I’m also not a physicist so it’s hard to dig too deep here without going out of my depth!
I’m getting Y2K vibes from the cryptographic issue. When I first learned of the risk many years ago, I was fairly alarmed at the implications. And, as it turns out, so were lots of other very...
I’m getting Y2K vibes from the cryptographic issue. When I first learned of the risk many years ago, I was fairly alarmed at the implications. And, as it turns out, so were lots of other very smart people who have been addressing the problem in the intervening time. So when quantum computers are eventually ready for prime time, the crypto thing will seem like an underwhelming nothing-burger — not because it was never a real problem, but because of all the groundwork that was laid in advance.
I'm no expert, but I see this as a snail race: we are going to be seeing announcements about progress in quantum computing for years to come, but it won't be practical. Meanwhile, the world is...
I'm no expert, but I see this as a snail race: we are going to be seeing announcements about progress in quantum computing for years to come, but it won't be practical. Meanwhile, the world is slowly moving to post-quantum cryptography.
Even when it's something that runs in a data center that you could theoretically buy, the algorithms that can be sped up by quantum computing are pretty specialized and my guess is that most organizations won't have a need to run jobs on these computers?
(That is, organizations other than governments cracking data encrypted with obsolete cryptography.)
So I see it as very prestigious research, but it doesn't seem much like AI in its impact on the world.
(Unless, of course, there's some important algorithm I hadn't heard of.)
People have little to no understanding of physics from the last hundred years or so and nobody cares. If quantum computer becomes commoditized people will just see it as the next cool thing to...
On the other hand, I feel that humanity is already struggling to keep up. I feel a sentiment of "just because we can, do that mean we should?" It occurs to me that a 256-bit quantum computer could likely break a lot of modern encryption algorithms, and trying to get people to understand the quantum paradigm when most already struggle to grapple with classical computation could further alienate people from modern technology.
People have little to no understanding of physics from the last hundred years or so and nobody cares. If quantum computer becomes commoditized people will just see it as the next cool thing to spend money on.
My fear is that google becomes a single entity that controls most of the computer world, from chips to connectivity to user terminals. Kind of like the old ibm.
It's neat, but Google are (naturally) very overselling this. Crypto cracking isn't going to happen till we have at least distance 17 surface codes. And claims of quantum supremacy are sure...
It's neat, but Google are (naturally) very overselling this. Crypto cracking isn't going to happen till we have at least distance 17 surface codes. And claims of quantum supremacy are sure technically true. But their benchmark is a pointless computation, so no-one has bothered to optimise for classical computers. This is where classical computers have a huge advantage - decades (centuries, in some cases) of optimising algorithms. Which is how tensor networks are beating NISQ
My cousin sent me this today, and after reading the article I've got some mixed feelings.
On one hand, quantum computing is an incredible technology that seems like it can push us further forward. The blog casually mentions parallel universes, which is kind of wild to see from such a major company like Google. Quantum physics has always been an exciting concept for me and having it become 'real' and tangible is fascinating.
On the other hand, I feel that humanity is already struggling to keep up. I feel a sentiment of "just because we can, do that mean we should?" It occurs to me that a 256-bit quantum computer could likely break a lot of modern encryption algorithms, and trying to get people to understand the quantum paradigm when most already struggle to grapple with classical computation could further alienate people from modern technology.
What are your thoughts?
These are well-considered thoughts! I respect the depth you put into this.
I don't worry too much about codebreaking, for a bunch of reasons. Most serious symmetric cryptography (so, government stuff) is probably resilient to quantum; organizations are already implementing post-quantum algorithms; and generally speaking, breaking encryption in one place doesn't get you all that much - you need to be able to do it all over, and even with a very fast quantum computer, that will take a long time.
What I've never considered before is your suggestion that people are already alienated from technology and quantum (which is an order of magnitude more complicated will make it worse). This is a really interesting thought. Ultimately, what I wonder is whether anybody really understands modern technology anyway. The best of us maybe have mastery over one field, but that field is probably pretty narrow. (How many excellent software engineers know anything about UV lithography? For that matter, how many excellent software engineers know anything about writing code for hardware?)
From that standpoint, I'm not sure quantum computing will make things too much worse - but I do think that there are likely to be quite a few interesting benefits.
To address the main point of interest, I'll qualify myself by saying that I have a degree in Computer Science and at best a vague understanding of anything that happens below compiler level. I totally understand that the world of tech (perhaps science as a whole) is built on layers of abstraction, and at the same time I think that the world of open-source is hugely comforting for me. I like to know that I am depending on as few "black boxes" as possible. I don't understand UV lithography, but I know that I could if I wanted to. I cannot say the same thing about Windows or Squarespace or Instagram, so my philosophy is to suggest that people steer clear of them.
Personally, I spend a lot of my time analyzing social media and digital identity on the internet, and a big part of what I aspire to do is teaching people how to "own their identity" in the digital world. I think that basic digital literacy is hugely lacking, and I already struggle to concisely explain in simplest terms what a website is and how it works.
As an 'ambassador of technology', people ask me what programming languages are or how domain names work, and I can usually answer those questions sufficiently. I've got a handle on explaining LLMs now. I don't really understand the implications of quantum computing, however, so I would be at a loss about how to explain this article to someone in layperson terms.
In general, I would agree that nobody understands everything about modern science. However, I believe that there is a certain degree of fluency that can be attained through study, and that sort of fluency is what enables people to pursue autonomy in the world. I believe that autonomy and ownership is what helps people confront the feelings of alienation, and we are already at a point where people do not feel a real sense of ownership over their online presence. This post isn't really mine, for example, as it doesn't live on my computer. I might make a copy of it, but probably not.
Seeing this article has made me seriously consider the household quantum chip, and how that will fit in to this larger image that I'm painting. What will people use it for? How will it be 'sold' to us? Will quantum technology become a cornerstone of the next scientific paradigm? Will we be able to leverage concepts such as quantum entanglement to share information over vast distances?
At the same time, will we even know what to do with it? The technology we already have in our hands could be used to create worldwide support networks and address problems on a global scale, but instead we use it to drive corporate profit and cope with the profound loneliness of our modern day by dissociating into constructed realities. If we bring these same habits and attitudes into the quantum world, I'm concerned that these issues will also be amplified by an order of magnitude - and I don't want to imagine what that might look like.
I think you’re crossing into science fiction. Quantum computing will allow us to do certain types of calculations very quickly. That’s it.
Quantum entanglement cannot transmit information.
It can't transmit information faster than light, but quantum channels still seem like an active field of research! I think there's lots of stuff we have today that was once the realm of science fiction.
Complete sidebar, but UV lithography is pretty neat! Asianometry has done some excellent explainer videos around the topic (bouncing lasers off of molten tin droplets! So cool :D), and these days, old photolithography processes are within grasp of exceptionally interested hobbyists! You don’t even need to be the child of a wealthy Silicon Valley tech person to do so :3
The problem is not that symmetric encryption is quantum resistant it's that TLS key agreement today isn't and governments today are recording traffic with the intent to decode it later, which they will be able to do, by attacking key agreement and then using the recovered symmetric key on the rest of the stream. I guess I only hope my own traffic isn't interesting enough for anyone to bother with.
I'm not saying that this isn't happening in some limited instances, because it definitely is. But there are a ton of challenges associated with this.
Collection - practical: collecting all traffic worldwide ever is much harder than you think. You need to either secretly control key nodes or force other entities to do it for you. How would the NSA collect random internet traffic in Finland, for instance? Even if we think that they control every node worldwide (they don't), they'd then have to send copies of all this traffic back to their data centers, meaning that all internet traffic would be effectively doubled. I think someone would probably notice.
Collection - legal: I know this sounds maybe a little laughable, but most Western surveillance agencies are prohibited from collecting on their own citizens without a warrant. Again I'm not saying it never happens, but having to do this all secretly makes it an order of magnitude harder.
Storage: How could you possibly store this much data? Think of how much data is generated every day - petabytes and petabytes. You can't tell what most of it is (because it's all encrypted) so you need to store all of it. That's impossible.
Archiving and search: okay, let's say I'm wrong and the NSA is secretly storing all your data. How do they know it's yours? All they would have is 'public IP address 1 talks to public IP address 2,' and those change pretty frequently. It makes searching through this stuff absolutely impossible.
Fun bonus challenges: even encryption susceptible to quantum decryption often includes things like the double ratchet algorithm that forces you to decrypt every exchange one by one. So let's say your quantum computer can decrypt a WhatsApp message every microsecond. That's pretty good! ...but turns out messages are being generated ten times faster than you can decrypt them, so you just fall farther and farther behind. Rats.
Ultimately, you have something that's extremely difficult to collect, impossible to store in bulk, maddening to search through, ...and it's all illegal.
I'm not saying this never happens. But I am saying that there's absolutely no way that it's happening in bulk for regular people.
Btw one last thing on double ratchets: the security of key rotation / forward secrecy is predicated on the idea that the key agreement steps cannot be attacked. But if they are attacked and you continue to see the complete stream of messages then you can simply mirror the stream of operations either alice or bob is performing without having to perform the attack every time because you acquired one of the dh secret keys + seed material. Signal isn't magic.
Seems fine to me. Ultimately a quantum computer just runs specific algorithms at a different complexity class (ie “way, way faster”) than a classical computer can. One of those quantum algos is capable of solving some of our popular encryption algorithms, but ultimately this is a known problem with solutions in progress.
Try not to do anything illegal and encrypt evidence of it in the next twenty years, but otherwise, this isn’t a huge deal. Twenty years ago most wifi connections were insecure and no one encrypted their network traffic. The goal behind traffic encryption has always been to make it seamless enough that people will actually use it, not to have them understand it mathematically, I feel, so this will be no different.
Re. The parallel universe and AI stuff, afaik it’s a combination of philosophical interpretations of the mathematics underlying quantum mechanics and hype chasing. This person is legally obligated to mention AI at least twice in all public communications.
Yeah, as a physicist I'll just say, no, this google chip does not create parallel universes in any meaningful way.
Cool, thanks for chiming in. As someone who doesn't really know his stuff, my mental model is that parallel universes are not created or destroyed, but something like concurrent processes which we are constantly moving between based on observation. I imagine that quantum technology allows us to determine what the outcome of an observation will be in a small cross-section of higher dimensional space and select the one with a result that we're searching for, with more powerful quantum computers being able to cast a larger field. Is any of this provable, falsifiable, or addressed by current theories?
I’d note that MWI is increasingly controversial, if nothing else because it relies on the linearity of quantum mechanics and many attempts to reconcile gravity and quantum mechanics create non-linear systems.
But even in MWI, you don’t move between parallel worlds. You simply are in one world. When there is a superposition described by probability, what is unknown is what world you are in, not the result, based on your current information.
No, it just lets you do types of math very quickly.
MWI is like fundamentally untestable since a basic underlying principle is that each world cannot interact with other worlds.
"Try not to do anything illegal" is a very concerning statement for me, mostly because we are living in an era where fascist ideologies are hot and trending. Twenty years ago, there was no widespread attention economy and the internet was a much different place.
Anyways, I do actually agree with the Many Worlds hypothesis and I consider quantum technology to be something that could allow us to leverage higher dimensions than the ones our physical bodies occupy. What we would do with that, I still don't know! But I hope we can direct it towards making this world a better place for all the beings that inhabit it.
Certainly, and agreed! But that said, if we assume for a future fascist dictatorship, evidence of misgivings will be unnecessary for arbitrary punishment. “Try not to record yourself breaking the law” is already a tough bar for many folks to clear (lighter side: people doing ‘social experiments’ on YouTube/TikTok, darker side: existing as LGBTQ+ and being doxxed), so I doubt that breaking encryption will be directly importantly for the overwhelming majority of us. Maybe it’ll matter more for three letter agencies.
I’m more of a pilot wave theory person myself, but I can see the appeal behind the other interpretations! I’m also not a physicist so it’s hard to dig too deep here without going out of my depth!
I’m getting Y2K vibes from the cryptographic issue. When I first learned of the risk many years ago, I was fairly alarmed at the implications. And, as it turns out, so were lots of other very smart people who have been addressing the problem in the intervening time. So when quantum computers are eventually ready for prime time, the crypto thing will seem like an underwhelming nothing-burger — not because it was never a real problem, but because of all the groundwork that was laid in advance.
I'm no expert, but I see this as a snail race: we are going to be seeing announcements about progress in quantum computing for years to come, but it won't be practical. Meanwhile, the world is slowly moving to post-quantum cryptography.
Even when it's something that runs in a data center that you could theoretically buy, the algorithms that can be sped up by quantum computing are pretty specialized and my guess is that most organizations won't have a need to run jobs on these computers?
(That is, organizations other than governments cracking data encrypted with obsolete cryptography.)
So I see it as very prestigious research, but it doesn't seem much like AI in its impact on the world.
(Unless, of course, there's some important algorithm I hadn't heard of.)
People have little to no understanding of physics from the last hundred years or so and nobody cares. If quantum computer becomes commoditized people will just see it as the next cool thing to spend money on.
My fear is that google becomes a single entity that controls most of the computer world, from chips to connectivity to user terminals. Kind of like the old ibm.
It's neat, but Google are (naturally) very overselling this. Crypto cracking isn't going to happen till we have at least distance 17 surface codes. And claims of quantum supremacy are sure technically true. But their benchmark is a pointless computation, so no-one has bothered to optimise for classical computers. This is where classical computers have a huge advantage - decades (centuries, in some cases) of optimising algorithms. Which is how tensor networks are beating NISQ