If someone cracked Bitcoin, why would they publish a press release? What's to market? Just make the money directly. It's like selling bitcoin mining machines. If they could possibly make money (depending on $/kW), it makes no sense to sell them.

There are reasons to publish a press release. Maybe you think Bitcoin (and other POW mining) is destroying the environment and you want to shut down the entire ecosystem instead of profiting from it. Maybe you prefer to be famous for an intellectual achievement instead of rich. And if you publicly cracked SHA256, I'm sure you can get a job that pays pretty well working on other projects.

Now, if your goal is maximize wealth, then I don't see a reason to publish a press release.

Exactly. The proof they did not actually crack Bitcoin is the fact they said they did.

Ag. The Satoshi Maneuver. I like it!

They didn't publish a press release. They filed for patents.

I will definitely hold my breath until these "patents" are granted. I'm just wondering how fraud and SEO spam are protected as intellectual property.

"Company claims they can beat the market. Just needs a small up-front investment to start winning big."

Clearly if this company had a way to generate bitcoins more cheaply than anyone else, they'd just mine the bitcoins.

How do you know they haven't been mining bitcoin? They've filed for 2 UK patents, which entails public disclosure of their methods.

From https://quantumblockchaintechnologies.co.uk :

"Quantum’s R&D will focus on Cryptography; bringing together the most advanced implementation techniques and functions, along with quantum computing technologies and AI deep learning, to develop a new and disruptive approach to blockchain technology."

This is very close to what I'd write as a parody. I don't think I've ever seen anyone unironically using "quantum computing", "AI deep learning", and "blockchain" all in the same sentence.

I have... Just last week, I saw on the news a still of an aspiring university rector giving a talk where all those terms (and "big data") appeared on the slide. Sadly I don't have any context because the news only talked about how he had been accused of running a fake paper mill. Unbelievable!

A company called "Quantum Blockchain Technologies" claims they use "Artificial Intelligence" to speed up Bitcoin mining.

You can't make this up.

Poking around their website makes my bullshit detector go off like crazy. The way they talk about machine learning speeding up hash search just makes no sense. And on archive.org you can see they have been talking like this for quite a while now.

Either they 1) cracked sha256, or they 2) proved P==NP.

Or 3) they've crammed every hype buzzword into their SEO blogspam website they could and will soon change their name to "Quantum Blockchain AI .com Ltd."

Quantum rainbow tables... now in 5D!

Seriously, this is 99.9% hype, probably generated by AI. I'm just waiting until the SEO spam salespeople flood the internet with Mark V. Shaney's that are entirely self-replicating automata of shit.

What are you talking about??? We already know P=NP!! [1]

In all seriousness, I'm enjoying musing about what kind of "fake it 'til you make it" group fever dream got them to think this would fly. Also can't wait for the follow up "damage control" press release that claims Satoshi is one of their principle scientists on staff.

Bonus points: That video of 'Method B' [2], --if it was real-- quickly decaying in advantage from 7x to 2.5x (as compared to brute-force enumeration) implies their patent-pending method is asymptotically exponential-time and that their heuristics aren't always that clever... yay slightly faster exponential time... sometimes! So I suppose they are being honest?

[1]: When N=1 or N=0 ■

[2]: https://youtu.be/vPT0HWVnE98?feature=shared

Thanks for sending me that. I checked out the rest of the channel. The CEO (lol) is in most of the videos, one of them is a heavily edited video where he says that his company not once, not twice, but 3 times independently found patterns in sha256, basically saying they cracked it, 3 independent teams all cracked it separately with different approaches.

No I don't for a second think they're being honest.

Are they offloading cloud operations using sidechannel funnel?

If so its a prime opportunity to look into investment potential and growth vectors!

> In general terms, I would think it’s possible to not calculate the entire hash to see if it starts with the required numbers of 0 for bitcoin but to break early and thus save time/energy.

... no? That's the point of the hash function, that you can't compute it without computing all of it. This statement is equivalent to breaking the hash function (very severely), which lots of thought has gone into.

Granted, if you do have a break for SHA-256, arguably one of the best ways to commercialize it would be to secretly mine Bitcoin much faster than others.

You can avoid all operations (actually, gates since we're talking ASIC circuits) of the last round that only affect the least significant bits (e.g. the last 176 bits if the threshold is 2^80).

Crypto-noob here: let's say they broke SHA-256. Even then, they'd need to know the hash in advance, which they don't, right? It's really hard for me to follow, I'm getting old :(

EDIT: Looks like no, they don't need to know the hash. Rasulek wrote a great overview here: https://news.ycombinator.com/item?id=39818305

I'm still confused about the leading zeros thing. I thought the article author was using a leaky abstraction. It sounds suspicious/notgood that you have a significant amount of info up front. But I'm just a noob free-associating.

Isn't this just to say that they cracked SHA-256?

Seems they're claiming they can find a predictable output from the hash function.

For those unaware, a cryptographic hash function should ideally take an input of arbitrary size and produce a fixed size output with the following characteristics:

- even distribution across key space - non-trivial predictability of input from output - completely uncorrelated output from input

It's the last feature that bitcoin relies on, which is that we need to take a given input (the block details) and come up with a random bit of information (nonce) to add to it such that the resulting hash output has certain arbitrary characteristics (i.e. begins with n number of 0s).

If they've found a way to make a hashes output more predictable such that you could determine an input that is likely to produce a given output, they could take a block, a randomly selected output with the proper characteristics, and find the appropriate nonce so as to achieve that hash output.

> Quantum Blockchain told FTAV it has not published any papers on technology or methodology “in order to protect their research”. Its patent applications are “as far as they have dared to go with making the developments public,” said a spokesman.

> That policy leaves a lot of questions unanswered.

The other way to prove that is to just start mining. People will notice immediately.

Funnily enough, I know about of people that believe easily provable but unproved claims like that. They don't even have to be as gullible as Craig Wright's followers, its seemingly a generational thing for people with some unrelated clout in one domain to be therefore believed in pie in the sky things. Especially in Europe for some reason, but I've seen it amongst Baby Boomers here in the states. Don't know how the rest of the world is.

I didn't realize Craig Wright had followers; I thought he's just a nutter shouting in the corner and everyone else rolling their eyes, like that fella who have been claiming he invented email.

I do wonder how many of them are genuine and how many are trolls just edging him on for teh lulz.

"Quantum Blockchain’s announcement that it can find these patterns with predictive AI also raised some doubts. "

Why is the FT even covering this obviously scammy company's claims?