One can visit this page to see the cloud-based quantum computing resources available now.
https://en.wikipedia.org/wiki/Cloud-based_quantum_computing
I would say there have been breakthroughs already and as with any new technology, it is very hard to determine its future usefulness and developmental timeframe this early on (especially if you believe some of the claims about computational power.) The use of AI and traditional supercomputers to aid in the design of quantum computers is another unknown variable. And besides quantum computers there are other 'orders of magnitude' advances being researched in traditional computing such as the emulation of biological neural networks.
Will bitcoin as implemented be useful for the generational transfer of wealth? I think it is something for long-horizon bitcoiners to consider. I am sure there are many people who have no idea there could be such future risks. If a first supercomputer hack occurs, it will be such big news. We would see a drop faster than that FTT token, I would think.
As discussed earlier, any supercomputer capable of breaking Bitcoin is capable of breaking just about anything and everything. A bad actor could wreak havoc right across the board. However, the chance of a fully-formed Bitcoin/everything breaker suddenly plugging in without warning seems remote in the greatest extreme. Bitcoin Miners and Bitcoin people in general have great interest in any and all technolgical developments that might affect Bitcoin. If just one Bitcoiner, or even just one person with a conscience, got to hear about it . . .
I'm as concerned about this scenario as I am concerned about NKorea suddenly, out of the blue, launching a new weapon that's so far advanced that the rest of the world is defenceless against it. It's theoretically possible, but . . .
Most crypto related anomalies like "Bitcoin Sender Struck With $3.1M Transaction Fee"; are related to the exchanges or simply due to various very unforgiving operator error.
https://web3isgoinggreat.com/
The 'unforgiving' side of things can seem scary at first but it's part and parcel of 'immutable'. It's nothing new - we all know that physical reality/time is immutable, and we're all generally comfortable with it. We know our lives are on the line when we cross the road, so we take great care. We should take similar care transacting Bitcoin, when our money is on the line.
I've thought about the super-fast encryption breaker problem too, because some outcomes might put an expiration date on our entire financial system.
The simple solution for username/password or username/biometric logins (i.e bank or brokerage accounts) is to have a time delay for each username's subsequent password attempt. A ten or twenty second delay, or a delay that gets longer with each incorrect password attempt, can extend the time it takes a brute force attack to locate a password to many years of trying random combinations.
This simple method is immune to the increasing speed of the hacker's equipment. However an AI information skimmer could do better by generating a few million potential passwords out of your personal information revealed online - year of birth, favorite things, favorite authors or characters, school mascot, sayings, word patterns, old passwords obtained from previous hacks, etc, thus generating a smaller solution set and guessing many password within a few months of trying, even with the time delays. An AI could also screen out possible passphrases that are not pronounceable, dramatically limiting the solution set.
With biometrics, there could be genetic patterns that could be assumed from information revealed online that reduce the solution set. An AI might find all sorts of patterns to make itself a better password cracker - patterns our sciences have never discovered. Maybe blue eyed people with type AB blood never have a certain pattern in their retinas?
So account-level time delays are a useful way to keep brute force hackers out of individual accounts, but they don't stop the hacker from trying a new username with each password combination in an attempt to breach any one account in the system. In large systems with potentially billions of users, the odds increase for the cracking program to randomly luck into at least one correct username/password combination, even if the odds are astronomical. The defender cannot set the whole system to time delay with each wrong password guess or else it could be locked up by anyone and thus would always be locked. If nothing comes up on the first attempt of a billion unique usernames plus a billion possible passwords, the hacker can roll again after the 10-20 second delay has passed, and so on. Even with an escalating time delay, the hacker has a reasonable chance of getting into
somebody's account because they can throw very large numbers at very steep odds. Being that somebody or reimbursing that somebody may be a cost of doing business.
At its core, the problem is our authentication data must be small enough to be convenient. The size of "convenient" means memorable in a human brain (containing dozens of other passwords) or portable as part of a human body (but then unchangeable if breached). The size of "convenient" doesn't change much, but the capabilities of brute force or AI-led password guessing attempts are constantly increasing. In theory, this dynamic leads to a crossover point where hacking capabilities exceed the possible security level of any convenient form of authentication data.
At that time we would no longer have a reliable way to trade in electronic formats, which could mean worldwide economic collapse if it happened fast enough. There is no guarantee offensive tech won't evolve faster than defensive tech, or the rules of our slow-moving economic systems.
Time delays could still be part of the answer, but at some point the brute force or AI-led attacks come so constantly that all user accounts are always at the maximum time delay. The maximum time delay can only be as high as a user is willing to tolerate. 20-30 seconds perhaps? One minute to make a sale?
Another possible solution is three-factor: username + passphrase + biometric, for example. I don't see this as a good solution because it's even less convenient and more costly than a time delay, and because it is essentially an extended password, which only kicks the can down the road. Plus the biometric element can never be changed, even if compromised. If we assume our data will eventually be compromised - as passwords currently are, then the biometric part is a false security blanket.
We could end up living in a world where we all carry around physical tokens containing authentication chains dozens of petabytes in size just to access each time-delayed account. Losing such a device might mean losing your money, unless some meatspace identity verification procedure was applied. Such a system would be far less efficient than our current world of 12-character-passwords-by-email. Yet it would get around the "convenient size" limitation and allow electronic commerce to continue despite quantum computing and ubiquitous AIs with near-infinite cryptographic cracking power.
We're currently using layers of cryptographic scrambling on as many of these elements as possible to prevent in-transit interception, but cryptography is essentially a problem of processing speeds and intelligence, and we can't hide behind these fast-falling barriers forever.
Cryptocurrencies deserve credit as an initial attempt to address this looming issue, but the method of hiding keys behind usernames, passwords, and website businesses built by shady characters in legal gray areas has been a disaster for so many people who have had their accounts and wallets hacked. If anything, all the false security of cryptocurrency only made it easier to steal. I bet there are technical lessons to be learned beneath the FOMO-throwing-money-at-obvious-frauds layer.
