Quantum computers and AI threaten data security. The only thing we can do is to be prepared for that threat.

"Researchers at the National Center for Supercomputing Applications are developing new cryptographic standards to protect against quantum computing threats. Their work includes measuring adoption rates and implementing quantum-resistant protocols, with early results indicating gradual progress." (ScitechDaily, Quantum Computing Threatens Cybersecurity: Are We Prepared?)

Mathematicians found the new prime number. (2^136,279,841 )− 1. That is a large number. If the computer uses Riemann's conjecture to break the code, that thing means that it takes a very long time to break the encryption that is made by using prime numbers. The system must try every known prime number for the message. 

That means traditional computers may take years to break the code. The neural networks can start simultaneous calculations at multiple points on a number line.  And that shortens the time very much the system can consist of thousands of personal computers. And that makes it possible to solve complex cryptological problems in a short time. 

The thing that connects neural networks to quantum computers is that neural networks might look like single, solid-core, or monolithic computers. But those networks are multiple computers. Those computers can operate independently. As well as. They can operate as an entirety. 

That thing means that when the neural network takes on a new mission, it must not stop. The traditional Turing machine must stop before it takes the new mission. If one unit of the neural network gets stuck, that means other participants will help it. That means the computer that is stuck can clean up its memories. And before it, the other computers can download that data into them. 

In the morphing neural networks, the system can analyze what caused the overleak. The morphing neural network means that the role of the participants changes. The AI-based control makes it possible to create a system with advanced self-diagnostics. That kind of system allows to use same systems for multiple uses. 

If the system also backs up the RAM- memory to independent hard disks. That denies the data loss if some units must be rebooted. Quantum computers can make similar things. It doesn't need to stop. But the quantum computers are new tools. And there is a long need to use binary computers. 

Three most powerful computing systems. 

1) PC-computer based neural networks 

2) Supercomputer-based neural networks 

3) Qauntum computer-based neural networks. 

Another thing is that quantum computers are coming. Maybe the NSA already has those systems. But also Chinese and their company states have their quantum computer projects. Those nations have money for that kind of system development. 

And that means our data security is endangered. The major problem is that things like RSA encryption are made to protect information against single computers. The major problem is that the RSA encryption algorithms are created for binary computers. It took years to create prime numbers using supercomputers. 

The quantum computer can create those prime numbers in less than a second. And that makes them the ultimate problematic tools. Those systems can also create an ultimate defense. The AI-controlled neural networks can also make it possible to create prime numbers more effectively than single computers. 

A supercomputer is a powerful tool, but neural networks are more powerful, especially in cases where the system must create the prime numbers from the number line. When somebody asks what is the most powerful computer system in the world, I would answer that the neural network of quantum computers. 

https://scitechdaily.com/quantum-computing-threatens-cybersecurity-are-we-prepared/

Networks make AI real.

Linux enthusiast Linus Thorwalds said that 80% of AI is marketing, and 20% is true. That is the normal relation in computer applications. The news about "doomsday AI": or, AI that destroys the world, makes AI look alike more fascinating than it is. But the fact is that the AI and its ability to generate the code should cause re-estimation in data security. 

The neural network can attack systems with powerful methods. In neural networks. The AI can change the attacking computers. And their IP's all the time. The AI-based systems also make it possible to use dynamic IPs that make it harder to track those computers. The large-scale neural networks can also operate as the artificial general intelligence, AGI. 

The news about the AI's ability to make code and even essays brings new users to those systems. Also, things like weapons and other kinds of stuff make the AI news interesting. Networks are things that make the AI real. They allow the AI to search data from the net.  Without them, the developers must program everything to the computer's memories. 

Or it makes it possible to generate a language model that can search data from the network. The data that the AI can use limits its abilities. Another thing that determines AI and its skills is the ability to operate in the physical world. The robot is the tool that turns the real actor who can clean the house and make food. 

Above:) Neural network. 

The fact is this. If we talk about things like AI doctors or AIs that make doctoral theses, we are far away from those things. There have always been people who cheat in universities. And the AI is only one tool for that thing. AI is the tool that makes many things more effective. But it requires that the user knows about the topics. The AI is an ultimate tool for people, who know what they do. 

The AI is a tool that can make many things easier. It can handle well-sorted and well-presented information like numeric calculations with very high accuracy. In a neural network, the AI can swap the calculation to another computer if that computer is required for some other mission. When a user takes the workstation in use, the system can swap the ghost or background process to another workstation. 

In that kind of network-based system, the system is based on the network of workstations. In the network-based solution, the single workstation can try to solve problems in a certain time. If the problem is not solved the system calls other computers to help. In this case, the system can scale the job theoretically unlimitedly. The only thing that limits the number of computers that the AI can use is the number of workstations and supercomputers in that network. 

But the AI cannot make things spontaneously. It always requires humans to begin the operation. 

Social media makes the AI more well-known. But AI is also a tool that requires social media. The AI companies use data from internet services like X, and Facebook to develop and train the AI. When R&D people make some new solutions for AI they require words that activate the process. In cases of the AI the word that people say or write causes a process where the AI searches for match. The word that the AI gets is the trigger that activates the process. 

The AI is still a regular computer program. Or, actually, it's a group of computer programs and applications. The marketing people call a language model that can connect itself to datasets as AI. Each dataset gives some "skill" to AI. The artificial general intelligence AGI is a large number of datasets that it can use to control things like robot taxis and microwave ovens. Otherwise, we can see a large group of dataset modules as one entirety. 

Mesh protocol. 

That means when the LLM calls the robot taxi, it just makes a "phone call" to the LLM that operates the taxi. Then the customer's LLM fills the form. And then it delivers responsibility to the AI that controls the taxi. The appearance of the AGI is formed when the cab transforms the voice it uses with the customer similar to the customer's LLM. There are two independently-operating AI-based systems run on two servers. The network-based structure that swaps missions between systems makes it possible to create a system that seems to be one entirety. 

In this case, the AGI is the network of independent operating systems that humans can control through the LLM. The network-based systems that use open architecture, or partially connected mesh protocols allow to connection unlimited number of modules to that entirety. And that brings AGI closer than we think. 

These kinds of network-based systems can have billions of full mesh networks and in the middle of those systems is the LLM. So each ball in the neural network involves LLM and the mesh network. Those systems can also have independent operating supercomputers. 

The thing is that the network of AI-based systems that connect robots with AI and LLM are impressive tools. The fact is that those robots are good marketing tools. People want to see robots in action. And that brings a new audience to the AI.