Topic: How long will existing encryption last? - page 3. (Read 2214 times)

As far as I'm aware, current attempts at making Bitcoin and other cryptocurrencies quantum-computer proof actually rely on using a complete different solution, rather than simply moving one step ahead by using more robust encryption.

See the following: https://royalsocietypublishing.org/doi/pdf/10.1098/rsos.180410

It outlines a slow defence against quantum attacks, and a novel solution IMO;

"We then propose a simple but slow commit–delay–reveal protocol, which allows users to securely move their funds from old (non-quantumresistant) outputs to those adhering to a quantum-resistant digital signature scheme."

Any symmetric system or block cipher using 128 bits or more won't be breakable by brute force anytime in the near future, and unlikely to be cracked within the next several decades or centuries. So using 256 or more is not a problem.

It's the asymmetric system, such as RSA and ECC which you are really concerned about. Still, 2048 or higher RSA public keys are unlikely to be cracked any time soon, so those using 4096 shouldn't see a problem either. ECC, I'm not so sure about, but whatever is the equivalent in length, same thing.

-
There is no connection between the problem of finding a private key knowing the public key (and knowing the ciphertext, the problem of asymmetric encryption systems) and the problem of finding the law of obtaining hash from input information.
These are fundamentally different problems. Mathematically, the first task is cryptanalysis.
The second one is solved by brute force attack.
That's why we need so many kubits to solve the SHA256 task.
But protection against this danger is also very simple; SHA256 can be easily made into SHA512 or 1024.
But you can't do that with a pair of public key + private key. This cryptography breaks down with any length of key, including mathematical.
For this reason they are looking for new post quantum systems to replace asymmetric encryption systems.
It is for this reason that symmetric encryption systems do not need to be upgraded to withstand quantum computers.

There is one more option, you can see the details here:https://bitcointalk.org/index.php?topic=5204368.0

If that is the case, I highly doubt it would be possible, because algorithms run by quantum computers are totally different, if they tend to break the encryption of bitcoin, they need to use the same algorithm that classical computers use, but with a bigger processing power, but who knows about it, I highly believe that even before a 2,500 qubits of quantum computing power would be invented, quantum computers do already generate a whole new set of encryption that will make it harder for quantum computers itself to break.

I read earlier today that it would take approximately 2,500 qubits of quantum processing power to successfully break the encryption of an SHA-256 private key.

Since Google only has a 72 qubit Q-computer, and it has taken a decade to reach this point, then a 2,500 qubit quantum processor appears to be approximately 7 years away.

With that said, this will still likely be a super specific system, so I doubt it would actually be used to identify the links between public and private keys.

In the meantime, no security system can withstand an attack.

They steal passwords, keys, confidential information in the most secure and well-funded security system - the banking system.

According to CNBC, the largest British retail banks have been forced to stop processing orders in foreign currency after a cyber attack on the exchange provider Travelex.

The company's computer systems were unavailable for more than a week after the malware attack on New Year's Eve, which left Lloyds, Barclays, HSBC and Royal Bank of Scotland, among others, unable to process transactions.

Ever stop the rise in cybercrimes?

There's something wrong with our "protectors"...

---------------------------
Unfortunately, quantum computers do not make encryption better.
Cryptography (encryption) is getting ready to accept the challenge of quantum computing.
But methods to counter such a threat are new post-quantum encryption systems on a regular computer.
The main requirement for such systems is the lack of modern encryption principles used in public and private key systems. Namely, factorization and discrete logirification in the fields of very large numbers.
And most importantly - no elliptical curves.
The reason for rejecting any cryptography on elliptic curves is that I have described more than once and in detail. The main reason for the unreliability of cryptography on elliptic curves is the elliptic curves themselves. There is no proven means to verify their safety. On the contrary, there is a lot of data about weaknesses in curves certified by the same NIST. And this is not a joke. This is like a special operation. Those who do not know about this problem, but do not know the absolute majority, use it. And those who implemented this cryptography use their knowledge, the weaknesses of this encryption method for their own purposes.

Many weaknesses of modern cryptography systems constitute a state secret. That is why this cryptography is not used in serious matters.

In everyday life - please encrypt, for those who really need it - they will decrypt it, but they will never tell you about it, this is a secret.

If this were not so, then we would not be looking for new post-quantum encryption systems, but simply would increase the key length in existing ones.

Blockchain based on SHA256 and ECC - in doubt today. But I think that its reliable place is only SHA256.
The fact is that the AES-256 remains a post-quantum system. The reason is simple - a quantum computer, even of the second generation, never breaks a 256-bit key, because, in symmetric systems, all the values ​​of this key work. And in asymmetric systems - only a very small part of the whole set.
But a symmetric system does not break with mathematical methods (this is cryptanalysis), and all modern asymmetric ones, including our ECC, breaks. Therefore, to increase the key length in ECC - it makes no sense.

Therefore, it does not accept any asymmetric system based on modern encryption principles for the NIST contest.

And if the AES-256 remains, then the SHA256 remains, this is the same level of reliability. Moreover, if the system does not break mathematically (by cryptanalysis), then increasing the key length, any number of times is not a problem.

See, I personally don't have any accounts in all those sites that were mentioned. But if I did, I would use a unique username and password for each one.

Password reuse is the biggest problem and people are just lazy to use different ones for different sites.

However, that is a user problem, not a problem of the system. The problem of the sites is a separate matter and that's a security issue.

Is there someone who could testify if that conjecture is true? Or some resources that could tell? Because providing an exact year of time by 2030 is quite suspicious, though, I do really think that future technologies such as the quantum computers could help make the encryption of blockchain technologies even more secured and strong. But if it could be used to decrypt encryption, I think we could somehow say bitcoin's encryption which is the SHA256 is quite in danger.

And then again the mass theft and then the total sale of our user data.

Earlier on the black market Dream Market in darkness were put up for sale 617 million accounts stolen from users 16 compromised sites. For $20,000. in bitcoins anyone could buy 162 million compromised Dubsmash accounts, 151 million MyFitnessPal, 92 million MyHeritage, 41 million ShareThis, 28 million HauteLook, 25 million Animoto, 22 million EyeEm, 20 million 8fit, 18 million Whitepages, 16 million Fotolog, 15 million 500px, 11 million Armor Games, 8 million BookMate, 6 million CoffeeMeetsBagel, 1 million Artsy and 0.7 million DataCamp.

I think that whoever invests 20,000 bitcoins in this illegal deal will not do it just to destroy them.

Let's say our, your, my private data, passwords got into these databases. What are we supposed to do?  Change our passwords, our IDs.

It's the constant change of our numeric identifier that suggests passwordless authentication technology. This change is permanent and it doesn't matter if your secret data is stolen or not.
Agree that not all of the information about what is being reported reaches us.

So, why expect this if you can prevent such an outrage, even if every day they steal now, will not be able to use it.

This is the innovation and essence of the technology of variables but deterministic for only one package of information, digital user IDs.
This is the essence of passwordless technology of user authentication by the server, user authentication by the user, etc.

 And such authentification occurs necessarily in two directions, on other it is not possible.
This completely excludes phishing regardless of the level of preparation of the attacker and the degree of carelessness of the user.

In general, a leak of data, any private data, may result in compromising accounts through password mining based on stolen information or using a stolen password.
Same scheme of attack development - on key information with the same or greater consequences.
Fraudsters also think about our bank data and methods of finding them or information that allows to access bank card data, pin codes, etc. in the same way.

As we can see from these observations on the logic of swindlers' attack, the ultimate goal is password, key, pin code and other permanent user identifiers.

The main word in this last sentence is Persistent Identifiers. And it does not matter what these identifiers are. What matters is their main disadvantage - their permanent nature.

Here is some known information, think about it:

1. Unknown persons have published unencrypted email addresses and user passwords in the public domain.  Security researcher Bob Diachenko discovered an unsecured Elasticsearch database back on December 4 this year, but it was indexed by the BinaryEdge search engine and has been publicly available ever since.
The database contained 2.7 billion e-mail addresses and over 1 billion unencrypted passwords to them. Database analysis showed that most of the data is a leak put up for sale by a cybercriminal under the nickname DoubleFlag.

2. In 2019, there were more than 14 billion user data records in the public domain around the world!
This figure is twice as high as the user data leak in 2018.

Thought it over.
Now the questions:
1) Why so many?
It's so much that calculations show - it's probably almost all users on the planet! 
2) And why is it coming out?
3) And where is the certainty that we are not in these bases?
4) Who says anything that steals in this sphere is published?

Who knows, is silent.

I draw one conclusion - we're all in danger of cyber crimes.

It's for these reasons that we advocate variable user IDs that make no sense to steal and sell.

Or here's an example, it's not clear how a security vulnerability worked, but it seems to me that they got to the keys - passwords - and made it a crime:

Yesterday, on January 3, Chrome extension stole $16 thousand in cryptographic currency!

A user of Ledger Secure malicious extension for Chrome lost $16 thousand in ZCash encryption. As it later became known, this little-known extension was disguised as Ledger's popular crypto wallet - the latter's developers had already disavowed the malware in the Chrome Web Store.
It is claimed that the Ledger Secure extension sends a passphrase to a third party, which allowed the attackers to steal 600 ZCashes from the victim's account. This user, nicknamed hackedzec on his Twitter account, also specified that he entered the passphrase on his computer only once 2 years ago and that it was stored as a scanned document.
Which storage option contributed to the theft of the crypt currency from the wallet is still unknown. How exactly the extension got into Chrome's browser also remains a mystery, but it was discovered when hackedzec found an unknown file on your computer with links to your Ledger Secure Twitter account. The account simulates the official representation of the French company Ledger.

Earlier MyCrypto detected similar malicious software in the Chrome Web Store. The extension, called Shitcoin Wallet, was freely distributed in Google's directory and stole private keys and authorization data from various cryptographic exchanges such as Binance.

What a twist!
Now we can't even trust the monsters the whole system relies on!

Tell me, where is the solid ground in this sea of uncertainty?

I'll tell you where, but few people will believe it - in systems without passwords and keys.

A paradox?
I don't think so.
It's a rescue.

It is not clear what is the point in reliable cryptography if no one is going to break it directly, but its keys are stolen.

Indeed, according to the same logic, it is unclear what the meaning will be in post-quantum cryptography or even post-post-quantum if the same keys are stolen.

The same security system holes remain and operate, regardless of the level of complexity of the system.

Maybe existing cryptography will not live long because of problems related to keys and the human factor? And not because of quantum computers?

I give an example of exploiting vulnerabilities that have remained a mystery:

-In 2014, it became known about the activities of a highly professional cybercriminal group called Carbanak, specializing in attacks on banks. It is assumed that the group managed to withdraw a total of more than $ 1 billion from various banks - while other cybercriminal groups failed to surpass this result.

Among the most noteworthy events, it is worth mentioning the large-scale hacking of the Italian company Hacking Team, specializing in the development and sale of hacker tools to special services of various countries. As a result, cyber attacks stole more than 400 GB of corporate data, which subsequently ended up on the Web.

But this is an organization that worked for the government, special services, which itself knows how to steal anything and from anyone - it itself has suffered!

But this is a real paradox.
If they did this to them, then what can they do to us?

Didn't the fundamentals of existing security systems based on keys and passwords compromise themselves completely and irrevocably?

How many more examples should humanity have to get in order to understand the inconsistency of the cyber security solutions that we are offered.

I remain a committed follower of new keyless encryption technologies and passwordless authentication methods.

There is a similar in this project: https://toxic.chat

The most sophisticated security system, any security system based on keys and passwords is vulnerable in these very places.
Individually, each of us can arrange secure storage of keys and passwords. But overall, it doesn't work well.

Here's the news again! The price of our security has dropped to a record $6. That's how much the program to hack into our accounts in one of the forums.

Check out the full text:
"Ring and Amazon have been sued for hacking into IoT video surveillance cameras.

The lawsuit charges the companies with breach of contract, invasion of privacy (!), negligence, unfair enrichment, and violation of the California Unfair Competition Act "by misrepresenting security".

Interesting wording: "by misrepresenting security".

In the same way, it is possible to formulate a claim against almost all companies that release all the software.

It's a sober view of our security situation.

But Ring has refused to comment on this situation.

Recall that in mid-December, credentials for thousands of Amazon Ring camera owners were published on the Internet, as well as 3,672 email addresses, passwords, time zone information and names assigned to specific Ring cameras (such as "front door" or "kitchen"). It has also become known that cyber criminals have created special programs to hack into company devices. In one of the forums, the user offered a tool to pick up Ring.com credentials for $6.
Here is the price for password and key security systems.

And the following news shows that such systems flow like a hole in a boat:

Provider of "smart" devices Wyze has leaked data to 2.4 million customers.

Smart Device Provider Wyze confirmed the data leak from the server.
Information such as client email addresses for Wyze accounts, names assigned by users to security cameras, WiFi network SSIDs and Alexa voice helper tokens used to connect Wyze devices to Amazon devices were leaked to the network.

Yes, I understand that stealing passwords (or keys) is not literally breaking cryptography, but it is a measure of the unsuitability of such technologies in today's reality. Yes, we've learned how to attack. More successfully than 10 years ago. And the techniques of such attacks are constantly being improved.

I am convinced that real superiority over swindlers can be achieved only through the introduction of new keyless encryption technologies and authentication methods without using a password, by variable digital identifiers (we are not talking about biometric identifiers), stealing and reusing variable identifiers makes no sense.

-------------------
There's always been a problem with cryptography.
The story even 10-20 years ago tells us that.

There are problems in cryptography now, except for symmetric encryption systems.

The problems that we see in cryptography are much more serious than the problems that a quantum computer will create.

Just before the quantum computer, the problems were known to a narrow circle of people and only to special organizations.

The advent of quantum computing has added new problems, which are now readily shared with everyone, in order to hide the real problems in cryptography.

Information for thought, even a theoretical very large quantum computer will not do anything with the number 256 bits in a binary system. And if you increase the key length in AES to 512 bits - you can forget about any fantastic calculations at all.

And if you increase the key length in AES to 1024 bits, even the idea of quantum computing becomes ridiculous.

In this case, the load on calculations will increase only 2-4 times, which is not a problem.

And the key length in post quantum systems with the length of 32 000 bits is considered small at all. There are systems with the key length up to 1,000,000 bits.
So what?
Or do you think these systems are afraid of a quantum computer with that much key length?

Therefore, a quantum computer is a terrible "Halloween" for the uninformed about the present state of affairs in modern cryptography.

Modern asymmetric cryptography (the one that is being replaced) is a temporary phenomenon based on unproven hypotheses.

The same is true for the security of the block-chain technology, a precisely temporary phenomenon, precisely based on assumptions that cannot be verified.

Details here (second post of December 4):
https://bitcointalk.org/index.php?topic=5204368.40.   

Quantum computers have opened the door to a new technology. The old encryption methods will now remain simple. However, this technology has not yet become widespread. No problem until it becomes widespread. Malicious people now have more action

It's nothing to do with the technology. There is no cure for user incompetence except training them on how to use systems. They trust the wrong things, that's not the fault of technology.

I'm not arguing, just telling it as it is. Users who give their passwords to other people or entities other than the official website in question, well ... that's how they compromise their accounts.

Well, phishing can be left in question whose problem it is.
The whole world has put that responsibility on the user.
I strongly disagree with this, and I'm putting this responsibility on the security organizers.
There's no point in arguing.

But it makes sense to look at the root of the problem.

As none of the times I have pointed out that until the basis of modern security system changes - the reliability of any new security system will not change.

In other words, all the upgrades and sewing up of holes will not stop the appearance of new problems in a system with an unreliable basis.

An unreliable basis for all security systems is keys and passwords.

It's a bold statement, but it's thoughtful.  You need to look at the essence, not the form.

I'll give you a fresh example to defend your position.

 You are a user. There is a manufacturer. The manufacturer is in trouble. You use it without suspecting that there are vulnerabilities that affect the Intel Platform Trust (PTT) technology and STMicroelectronics' ST33 TPM chip.
What do you and I (users) have to do with this?
Well, here's the answer.
 Vulnerabilities in TPM chips allow stealing cryptographic keys. A team of researchers from the Worcester Polytechnic Institute (USA), the University of Luebeck (Germany) and the University of California at San Diego (USA) discovered two vulnerabilities in TPM processors. Exploiting problems commonly referred to as TPM-FAIL allows an attacker to steal cryptographic keys stored in the processors.
This chip is used in a wide variety of devices (from network equipment to cloud servers) and is one of the few processors that have received CommonCriteria (CC) EAL 4+ classification (comes with built-in protection against attacks through third-party channels).

The researchers have developed a number of attacks, which they call "timing leakage". The technique is that the attacker can determine the time difference when performing TPM repetitive operations, and "view" the data processed inside the protected processor. This technique can be used to extract 256-bit private keys in TPM used by certain digital signature schemes based on elliptical curve algorithms such as ECDSA and ECSchnorr. They are common digital signature schemes used in many modern cryptographically secure operations, such as establishing TLS connections, signing digital certificates and authorizing system logins.

So this is the subject of our disagreement - keys and stealing them.

It turns out, "A local attacker can recover an ECDSA key from an Intel fTPM in 4-20 minutes, depending on the access level. Attacks can also be carried out remotely on networks by recovering the VPN server authentication key in 5 hours," the researchers note.

This news would not be revealing to our discussion,
if news like this hadn't come from all over the world like rain.

There's no cybersecurity, it's a software salesman's myth.
Think of the number of critical updates released by Microsoft (or rather microscopic software) to their operating systems, exactly like a storm... 

--------------------
I fully agree with that opinion.
 But I do not agree that stealing password and other personal information by means of phishing is not a technical problem and it is the problem of inattentive user.
It's not just your opinion, it's a public opinion.
Moreover, I think this opinion has been softly imposed on society by those who cannot and do not want to solve this problem using technical methods.
I'm sure that society will change soon.
Phishing is possible only when you do not authenticate the website, but only the website authenticates you. Only with one-way authentication.
Moreover, once you are caught in phishing, you lose a lot, you do not know that you are already attacked, or you will never know about it.
The security system makes this problem our problem.
And I think it's an old, wrong opinion imposed on us.
I think it's technically possible to do two-way authentication.
 But there's more to it than that. We need to ban authentication with permanent identifiers, as it is now.
These technical measures will completely eliminate phishing as a method, as a phenomenon.
And instead, we are offered to "look closely" at the site and remember in detail how it looks.
This is in the 21st century! This is ridiculous! It means that the whole old security system is unsuitable in our time.
I recently read how phishing attackers deceive the most attentive users - they take high-quality photos of the site and put the necessary active windows to enter the login and password.
What to do in this case? To be very attentive is not a method, it is a complete failure of password authentication technology.

I notice that it is a problem, but I also notice that it is mostly, or almost all of it, a user problem. It's not a technical problem. Good password systems do work. It's the users that reuse old passwords, or increment digits to new passwords, or some other variation that is now found in most brute force password cracking algorithms.

People using names, using dates, using numbers that look like dates, using words they thought only they knew but are in many dictionaries, and using any of those previously mentioned mixed and matched up with rules that are now configurable in the latest generation word list generators for crackers.

Randomly generated passwords do not just get hacked. They are found through some other weak spot in the entire system. Usually, it's the user. Or some other low tech method like a hidden camera over the keyboard, or a hardware keylogger that can't be detected.

Anyone who is smart enough to use a password that looks like a Bitcoin address or Bitcoin private key, just once, for only one particular website or system, and uses completely different passwords for different systems, do not get hacked unless targeted individually by government agencies. Then you're screwed no matter what.