oooorrr - and hear me out - investments are inherently hype-based and irrational and there is too much money flying around to do actual smart decisions
Quantum Computing (QC) is unlike previous technologies which were all mostly "logical structures" (i.e. the underlying Physics/Technologies were well-known). The viability of both the core Physics itself and its realization through Technology for QC are questioned by some Physicists/Technologists themselves. But in 2024/2025 many Govts. and Companies both have started investing heavily in QC. Moreover the advanced countries have implemented export controls on QC technology prohibiting export of QC computers above 34-qubits.
And now the ACM prize for something done long ago in quantum information.
Finally note that QC algorithms can be simulated (for small size qubits) on conventional computers and the current AI technologies may also play a part here i.e. implement QC algorithms on the "Cloud supercomputer" and using AI technologies.
The logical inference is that there has been some technological (one or more) breakthrough in the realization of the QC qubits technologies, QC algorithms running efficiently on the cloud, AI usage for QC etc. Nothing else explains all of the above facts.
See also: The Case Against Quantum Computing by Mikhail Dyakonov (Professor of Physics) - https://spectrum.ieee.org/the-case-against-quantum-computing
aaand you entered the "hype and irrational" territory. I dare you to reread your own comment, it is funny
right now QC is 5 orders of magnitude away from practical systems - there's NO profit to invest for. It's all research that is being hyped and overpromised because there's not enough money in that sector and because established players (like google) don't want to lose their face
viability of core physics does not imply immediate creation of product. I'd point to fusion, but that's also currently getting over-hyped 15-20 years too early
governments are only investing the same way as into particle accelerators - in form of research grants
simulation of QC is both extremely trivial (in "exponentially-slower" way) and existentially impossible (the whole sector would not exist if it was actually possible to use good old normal CPUs fast enough). Bringing in "AI technologies" only shows you as a gullible idiot that still parrots ai bubble without understanding exact details
If there is a breakthrough - it is secret government information, and it would not be available to non-government companies, especially those you can invest into. The moment such breakthroughs reach the market, knowledge of the very existence spreads - and yet all current known progress is dull.
The only evidence worth anything out of what you brought up is the export controls - and those have been extremely pre-emptive in preparation for geopolitics and far future tech. Error-correction barely started to be useful at 100 cubits, so 34 makes no sense other than to minimize brain drain with base tech
You have not understood the first thing about what i had pointed out.
> right now QC is 5 orders of magnitude away from practical systems - there's NO profit to invest for. It's all research that is being hyped and overpromised because there's not enough money in that sector and because established players (like google) don't want to lose their face
While there has been hype, in the last couple of years things seem to have changed and now culminated in the awarding of the ACM Turing Award prize. Do you know anything about the Physics/Mathematics behind qubits (eg. probablities/superposition/phase/noise etc.) and/or how that has been realized via technologies (eg. superconducting/photonics/trapped-ions etc.)? People are looking at "hybrid" quantum computers i.e. conventional+quantum (eg. IBM, Fujitsu), shuttling qubits on silicon (eg. Hitachi) which allows existing foundry technology to be used for QC. This is huge.
> viability of core physics does not imply immediate creation of product. I'd point to fusion, but that's also currently getting over-hyped 15-20 years too early
Non-sequiteur.
> governments are only investing the same way as into particle accelerators - in form of research grants
No, Govts. are actively funding startups in this area and including technology research/transfers in their Free Trade Agreements with other govts.
> simulation of QC is both extremely trivial (in "exponentially-slower" way) and existentially impossible (the whole sector would not exist if it was actually possible to use good old normal CPUs fast enough).
Simulation of QC is not "extremely trivial" but requires HPC technology. Datacenter/Cloud technologies are also utilized here. Generally only around 30-50 qubits have been simulated with 50+ qubits being exponentially prohibitive in terms of compute power/memory.
>Bringing in "AI technologies" only shows you as a gullible idiot that still parrots ai bubble without understanding exact details
To use your own language; this right here shows that you are just a clueless idiot about this domain. AI is a tool applied to various domains eg. AlphaFold for protein structures in Biology which solved an almost intractable problem. People are doing the same with QC+AI. There are a bunch of papers on this; for your edification start with Quantum Computing and Artificial Intelligence: Status and Perspectives - https://arxiv.org/abs/2505.23860
> If there is a breakthrough - it is secret government information, and it would not be available to non-government companies, especially those you can invest into. The moment such breakthroughs reach the market, knowledge of the very existence spreads - and yet all current known progress is dull.
This demonstrates your gullibility. Since one of the best studied usecases for QC is cryptography, if there has been a breakthrough in some lab (govt/academia/company all of whom have secrets), the powers-that-be would not want it to be widespread for security (mainly) reasons. But hints might have been given and investments encouraged. Almost all QC companies have a govt. tie-up and cryptographic technologies have already been subject to export controls from the very beginning. Another scenario is defense applications. There are plenty more but these two are the main ones.
> The only evidence worth anything out of what you brought up is the export controls - and those have been extremely pre-emptive in preparation for geopolitics and far future tech. Error-correction barely started to be useful at 100 cubits, so 34 makes no sense other than to minimize brain drain with base tech
That is the obvious superficial take. Given what i have written above, what if semiconductor technology i.e. the "hybrid" QC+Conventional allows one to simulate 100+ qubits easily now? What if there has been some breakthrough's by using AI on QC algorithms both existing and new ones? Have any formerly intractable problems in Physics/Chemistry/Biology/Mathematics been made tractable now due to AI usage? How many of these can be implemented on a QC? Etc. Etc.
To summarize; you have to look at the whole complex picture before drawing conclusions. Merely parroting trivialities like "hype" is meaningless.
