hckrnws

If you want to try quantum vibecoding, I threw up a site at https://www.haiqu.org where you can mcp with the quantum computer at TU Delft. Free, after you make an account.

Incredibly cool initiative! Looks like they're going for a trapped-ion device, which the best you can get for now. It's not clear what kind of geometry the ions will be on, but I assume linear traps? If so, it can't be scaled beyond 10ish qubits, so that's definitely more of an educational project. That makes sense though, since the other options like racetrack or whatever are still active research.

I wonder if there's ever been any cross-pollinating between SC and trapped-ion labs when it comes to control electronics and such. This could be a good way to find out.

Couldn't find any cost estimate, but from https://openquantumdesign.org/the-quantum-computer (scroll down to "What's Inside") I'm guessing 100's of k$ for the bill of materials (let alone keeping the thing going.)

So the "you" in "your own" has to have pretty deep pockets...for a relatively low fidelity 30 qubit device.

> So the "you" in "your own" has to have pretty deep pockets...for a relatively low fidelity 30 qubit device.

Sure but once you buy it, all the "you"s in all the other universes get to have one too.

And yet they won't split the bill with me. Bunch of freeloaders.

Banking Company LLC presents: Quantum Loans™ "Your money is simultaneously yours and ours until you check your balance."

Superposition Financing: Your loan exists in all possible amounts until observed. Checking your balance collapses the wavefunction — so we recommend you simply... don't. Ignorance isn't just bliss, it's financially optimal.

Multiverse Co-signing: Split the debt across all versions of yourself in the multiverse. Sure, some of you will default — but statistically, infinite yous means infinite revenue for us.

Entangled Interest Rates: Your rate is entangled with a partner borrower chosen at random. If they pay on time, your rate drops.

Payment Clusters: Forget monthly installments. Payments arrive in probabilistic clusters — sometimes three in a week, sometimes none for a year. We can't predict when, and neither can you. It's not a bug, it's quantum mechanics.

Most of the cost is in overpriced laser systems; if that gets solved a trapped-ion system could be reproduced for few tens of thousands of USDs. Still not a hobby weekend project, but certainly more attainable for more universities.

Okay so not a weekend project.

For those interested in the compiler/software stack and control hardware: https://pennylane.ai/blog/2025/12/open-source-quantum-comput...

> But unlike commercial ventures, the plan is to give away all the intellectual property

I would prefer the term "share". "Give away" implies a single, proprietary enterprise could acquire and control it, which isn't "open source".

This effort is likely aimed at industrial/academic entities and not "you" as in a single person. But anyway, it needs to be emphasized that the phrase "quantum computer" is today used to mean anything ranging from

-a useless machine that produces a signal indistinguishable from noise TO -a highly sophisticated marvel of science and engineering that performs otherwise impossible calculations

Many industrial quantum computers today fall closer to the the former category than the latter. A single person or small team with minimal funding has basically no hope of building anything meaningful.

I don't know of any other device that has such a broad range of quality represented under one name. Maybe like calling ELIZA and Opus 4.6 both "AI".

All “industrial” quantum computers currently fall entirely in the former category. Anyone trying to tell you otherwise is selling snake oil.

Is there a QC out there that can perform a commercially useful computation? No, not yet. And yes snake oil is abound. But the reality is not two categories, it's a spectrum. Some are more useless than others.

No, it's not a spectrum in any meaningful sense. There are scam companies (some with semi-respectable research departments attached to them) and there are research projects. Anyone selling devices with the promise that those devices will do anything useful for their customers are simply lying.

It's like fusion energy: there are legitimate companies working on the problem, and they may even succeed at some point, but anyone willing to deliver a 1MW fusion plant tomorrow is scamming you, because the technology doesn't work yet.

I have far more faith that fusion someday might be useful than I do for quantum computing.

[dead]

The first QC that decrypts previously undecipherable text will have incalculable value to the government that surrounds it. QC companies are bullshit because they will take whatever free non-gov money they can, why not? They exist to absorb government money, rightly so, but their public profile is simply to get money from private sector sources

Do we have an example of a real quantum computer doing some kind of a computation that is not easily accessible by the regular computer?

I keep hearing about "the promise" and "achieving quantum supremacy" (again!), but is there a real example of a quantum machine doing something useful in real life?

No, there are none; the closest we currently have are various special purpose and more or less hard-coded machines that demonstrate that scaling exists; various general-purpose machines operating on handfuls of qubits demonstrating the various gates; and various snake oil scams that may or may not have semi-respectable research divisions associated with them.

Interesting, do you work in that field?

The Venn diagram of "useful" and "not possible on a classical computer" has demonstrations on both disjoint ends but is currently empty in the intersection. For now. I fully sympathize with the hype-fatigue though.

What about Schor’s algorithm?

That's on the useful end but I don't think any QC has gone beyond being able to factor 14 or something in that neighborhood. Realistically we'd need a few thousand qubits to factor anything that's reasonable and current QCs have a dozen or so qubits that work.

https://www.newscientist.com/article/mg26735582-700-how-a-we...

No. It's as if there's a "No computing with this shit" theorem, enforced by nature alongside "No FTL communications," "No hidden variables," "No, you can't build a transporter, not yours," and "No cloning."