Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Quantum computing bombshells that are not April Fools

scottaaronson.blog

Post reply on HN
Print view View original

111–118 of 118 posts

Re: Quantum computing bombshells that are not April Fools

#111

Post by eru »

tombert wrote: post #84
eru wrote: post #45

Earlier quoted context omitted.

You might benefit from https://en.wikipedia.org/wiki/Index_fund

Like 95% of my investing money ends up going to fairly-low-risk ETFs like VOO or VTI, not too different than from an index fund. Still, that last five percent is more or less my gambling money. I put it into individual stocks with the hope they get huge. Sometimes it works out well, like when I bought $700 of Nvidia in 2022. Sometimes it goes badly, like when I bought a bunch of Sears stock with the hope they'd bounc…

Well, if it comes out of your gambling budget and you treat it as entertainment, then there's nothing wrong with that. At least the expected average payout is way better most other forms of gambling.

Re: Quantum computing bombshells that are not April Fools

#112

Post by whattheheckheck »

freetonik wrote: post #57
sigmoid10 wrote: post #42

Earlier quoted context omitted.

If they're going public I imagine they already sell some kind of QC chips. But, like, who buys them? Yesterday there was a new paper [1] that shows how Shor's algorithm could break realistic encryptions with as little as 10,000 qubits (instead of millions), but as far as I know quantum hardware is still orders of magnitude below even that target. So how big can the market actually be? Shipping to universities or othe…

>There is zero profit to be made at the end of the chain. Think of all the "sales" that comprise such things as space missions (ones without immediate real-world use) or large hadron collider. Or any other large, expensive, long scientific project. If you measure the outcome purely in money within decades, these things can be said to be zero or negative profit. How much profit was at the end of the chain of the curre…

Profit is serves as a decentralized coordination for the allocation of scarce resources of the infinite number of commodities and combinatorics of products and services that exist in the world.

Have you read Basic Economics by Sowell?

Re: Quantum computing bombshells that are not April Fools

#113

Post by tsimionescu »

sMarsIntruder wrote: post #99
tsimionescu wrote: post #64

Earlier quoted context omitted.

The banking sector can relatively trivially move to a new encryption scheme - this is one of the huge advantages of centralized systems. Also, the banking sector, rather than trying to provide anonymity/pseudonimity, has KYC - they can relatively simply disable remote access if they think a current system is no longer secure, and get everyone to come in to a physical office and get a secure version, after manually ve…

Upgrading decades old legacy systems? Good luck with that.

External access to banks is already using the latest versions of TLS, so no luck needed, they will just need to upgrade as they do constantly.

Re: Quantum computing bombshells that are not April Fools

#114

Post by ProllyInfamous »

freetonik wrote: post #94
ProllyInfamous wrote: post #83

Earlier quoted context omitted.

Thank you for posting/sharing. My mid-sized US city (Chattanooga) has recently announced a partnership with Vanderbilt and EPB (local govt-owned fiber ISP) which creates a Quantum Computing Research Facility [$,$$$,$$$,$$$] [1]. As locals are covering this news, I keep having this thought that nobody (perhaps less than a few ?) even knows what those words mean (certainly not me) . You speak confidently and clearly en…

The meme refers to the notion where an observation (i.e. interaction) collapses the wave function to a single value. As in, prior to observation, a system in a quantum superposition is said to be "in multiple states at the same time", and after the obsevation only one state exists, while all other possibilities are gone (or exist in other worlds, according to one of the interpretations of quantum mechanics [0]). So,…

I'm about 20 minutes into:

>>https://www.preposterousuniverse.com/podcast/2024/05/13/275-...

Which was recommended from the top_commenter's link to his own professional QPU programming presentation. I remain interested in learning more about this pure magic.

Ya'lls words both seem to represent worlds which I doubt even more words could help me into ever understanding(s).

I definitely understand being "in multiple states at the same time" — — —

Re: Quantum computing bombshells that are not April Fools

#115

Post by geoffbp »

jb1991 wrote: post #82
geoffbp wrote: post #41

Earlier quoted context omitted.

Reading mode in your browser…?

I would love to use reading mode in my browser if this site supported it. As I noted.

If you request the desktop site, reading mode will be available (that’s my experience anyway).

Re: Quantum computing bombshells that are not April Fools

#116

Post by sigmoid10 »

freetonik wrote: post #57
sigmoid10 wrote: post #42

Earlier quoted context omitted.

If they're going public I imagine they already sell some kind of QC chips. But, like, who buys them? Yesterday there was a new paper [1] that shows how Shor's algorithm could break realistic encryptions with as little as 10,000 qubits (instead of millions), but as far as I know quantum hardware is still orders of magnitude below even that target. So how big can the market actually be? Shipping to universities or othe…

>There is zero profit to be made at the end of the chain. Think of all the "sales" that comprise such things as space missions (ones without immediate real-world use) or large hadron collider. Or any other large, expensive, long scientific project. If you measure the outcome purely in money within decades, these things can be said to be zero or negative profit. How much profit was at the end of the chain of the curre…

I don't think any of your comparisons are accurate. Artemis may be a publicity stunt and a jobs program by congress, but the companies are in it for the long term mission of commercializing space. And SpaceX has clearly shown there's already a lot of money to be made with commercial space programs.

The LHC/CERN on the other hand is officially a pure science project, fully funded by public institutions. Their goal is not to go IPO and raise money for more experiments from some vague promise of potential future commercial applications, even though they probably exist.

Re: Quantum computing bombshells that are not April Fools

#118

Post by socketcluster »

Nevermark wrote: post #39
socketcluster wrote: post #37

Earlier quoted context omitted.

Why would experts care about my product? There's no big money behind it. The big money has to come in first, then the experts come later to tell the big money whatever they want to hear. Maybe they want to hear the truth maybe not... Either way the paymaster always hears what they want. Besides, I am an expert. I studied cryptography at university as part of my degree. I have 15 years of experience as a software engi…

You are leaving something out then. Which you allude to. Bravo on five years! I recently solved a problem that took me over 30. I originally thought, 3-5 months maybe, then 3-5 years, ... I am happy it didn't take 50. I have killed a lot of my own darlings. Well apparently you know what you are doing, I am sure you have something. I have found the best language models are great at attacking things. You may have alrea…

I built my quantum-resistant signature scheme from the simplest cryptographic primitives I could find precisely to mitigate risk.

This implementation simplicity came with tradeoffs; I chose a stateful signature scheme (Lamport OTS with Merkle Signature Tree) which I could fully understand and easily verify from first principles. It's actually one of the simpler aspects of my blockchain.

Originally, I wanted to use SPHINCS+ (stateless) but there were no good libraries for it at the time in the Node.js ecosystem and TBH, I didn't fully grasp some of the finer technical details at the time; it was clear it would have been a much more complex and error-prone solution.

That said, statefulness in the signature scheme creates complexity elsewhere and shifts the risk to the clients. There are risks that a poorly implemented client could lead to key-reuse. Also, there are concurrency scenarios which can't be supported securely (e.g. if the user tries to sign multiple transactions in parallel from two different clients from the same wallet/private key).

And yes, I did run several versions of Claude including Claude 4.6 sonnet against both:

https://www.npmjs.com/package/lite-merkle https://www.npmjs.com/package/lite-lamport

It said it found 2 non-critical issues with lite-lamport but neither was exploitable.

I had to add comments to the README so that anyone who tries to run an AI against the code knows that I'm aware of these concerns but they are not issues. I did fix a non-critical signature malleability issue but that was more to appease the AI since there was no exploit for it at the level of the blockchain; it basically ignored any extra trailing bits at the end of a valid signature which could in theory lead to replay scenarios with an extended signature with junk data at the end but not possible in practice since transaction IDs are unique and deterministic. There's no possibility of double-spend.

I look forward to running the new Claude Mythos on it when it comes out. In the meantime, if anyone can find an issue with it, please let me know.

Post reply on HN