Modern hackers are like ghosts.

Hackers are like ghosts. They search networks and try to find a weakness. Those people can search for data. And they can sell it to people. Who are willing to pay. Governmental supported. Hackers are the new operators on the net. Those people are classified as “advanced persistent threat” (APT). The problem with government-supported hackers is this. Those people are working under the control of governments. 

They are untouchable. And they have official authority. Those people have the most advanced AI-based tools in their operations. And they might have very expensive development tools. Another problem with those people is this. Those hackers can work for organized crime, and organized crime can offer many things. To a government that has exportation limits. 

Those people. Can sell things. Like drones, software, and computers, to the North Korean government. And that the government can offer assault rifles to those criminals. The hacker who gets things. Like geospatial information. Like. Precise locations of nuclear warheads, or information about things like radar frequencies that radar satellites use, can cause large-scale damage. When the data layer. Between. Humans and the environment spread. 

Hackers can become. More and more dangerous. When the AI connects physical layers. Like robots and 3D printers on the net. That causes problems for physical security. The misuse and misunderstandings can also cause trouble. When we think about things like a blood-type table, it’s possible that a hacker accidentally presses the “delete” button. Or, otherwise accidentally. Removes vital information. 

A modern hacker is not a 15-year-old kid. Those people can work for organized crime, or governments, or both. Those people. Those. Who is operating behind hackers can force people work for them. They can threaten their life or their family members. 

The fact is that if the system exists, it's vulnerable. And things like social media. That is open social media. And restricted social media. Like company meetings. Are platforms that allow hackers hunt people. The virtual actor can be used as a phishing tool. 

Those actors can slip into the company meetings. And the problem is this: remote meetings are on weekdays. In multinational corporations. Remote work is not itself a problem. It’s possible that the workers must travel to their offices. Otherwise, there are no safe offices. Things. Bluetooth earphones can act as eavesdropping tools. The eavesdropper can sit just behind the wall and use that equipment for spying. 

The vulnerability is not always direct. It can be indirect. Indirect means that there can be a vulnerable surveillance camera in the office. And if hackers can slip into that camera. They can read things. Those are written to monitors. Or they can make their own key to those offices, and then they can search for papers there are passwords. Or include their own telephone. To the micro support’s service list. So, they can simply ask. For their passwords in the targeted system. So, only one unprotected telephone can open access to the system. 

Those earphones can be on the table or the shelf. The eavesdropper can use the same tools that police use for sound analysis. And that turns even the lowest noise into understandable speech. And the speech-to-text applications make those things more effective. The hacker can read what people say on the screen. And that helps to destroy the privacy. And how can a hacker open the door? They can break into the property maintenance company’s computers. And include themselves. Into the list of people whom those workers must open the door for. 

But. The problems are the overseas meetings. Those meetings where people make big decisions can be thought of as safe. But if somebody can slip a virtual actor into those meetings, people might not worry about that possibility. The thing is that hacking itself doesn’t require programming skills. It requires the ability to get things like passwords. 

The problem is this. While workers are in offices, their homes are empty. The attacker can slip into those homes and try to find things like notebooks. And there are those passwords. When we think that we are safe, we are under the biggest threats. Things like blackmail are tools that open the route to the system. The wrong type of trust can cause big damage. 

Drones are the most considerable new weapon systems.

Above: MQ-28 "Ghost Bat" launches AMRAAM. These types of drones. It can also carry air-to-ground missiles and anti-radiation missiles. Like. AGM-88 HARM and other smaller drones. 

Drones are a new and effective threat. The reason why drones are so hard for defense is that their development cycle is much faster than that of any manned systems. Drones are cheap to build. And their operators' training. It is fast. And cheap. A drone doesn’t need life support systems. 3D printing systems. Make it possible for drones. To be customized for each mission. 3D printing systems can be installed in ships or trucks. And they can create drones from plastic, carbon fiber, or metal wires. This means those drones require only microchips. And engines. And explosives to make effective surprise attacks. 

The fact is that drones don’t need any explosives to be dangerous. They can roll metal wires around high-voltage power lines. Those systems can also have Kevlar whips. They have a metal bite at their end. And those whips can hit targets with supersonic speed. The Shahed-136 and Geran-3 type drones can strike targets from 1000 km distances. Those drones are used to drop mines in Ukraine. Those drones can also drop other drones whose mission is to cause damage and suppress air defense. The AI-based programming makes those drones very independent. And that makes it possible. Those drones are left behind enemy lines. And then they can be activated remotely. This means those drones can sit and wait for their orders. A drone can attack things like submerged targets. This means they can be dangerous. To full-size submarines. And it's possible. 

That the full-size nuclear submarines can be converted into sea drones. A submarine’s nuclear reactors. Can deliver energy. For. A full-size supercomputer. This type of system can drive complex algorithms. Things like torpedo tubes are easy to convert for automatic reload. Torpedoes and missiles can be stored in long tubes. The system mimics the firestorm system, but it uses torpedoes. And that allows the submarine to launch them at a very high rate. Large-sized submarine drones can carry large nuclear warheads. And that means. The system. Can carry. 4-5 “Tsar bomba”-type warheads (50mt.) along with torpedoes and missiles. That means the Russian “Poseidon” system can be a child’s game. Compared to systems that are coming. 

"China’s heavyweight jet-powered Jiutian drone, said to have a maximum takeoff weight of around 17.6 tons (16 metric tons), has flown. A key mission for the design is expected to be acting as a mothership for swarms of smaller uncrewed aerial systems, as TWZ has explored in the past. It has also been shown previously armed with various air-to-surface and air-to-air munitions, and could perform a variety of other missions, including airborne signal relay and logistics." (TWZ.com, China’s High-Flying Swarm Mothership Drone Has Flown)

Things like large-sized drone motherships are also coming into service. Those systems. It can be a new threat and system for dominating battlefields. Those systems can carry kamikaze-drones. But also UCAVs that can operate against other aerial targets. 

Aerial drones. They are classified as unmanned aerial vehicles (UAVs) and unmanned combat aerial vehicles (UCAV). The difference between those drones is not big. Things like kamikaze-drones. Can attack both ground and aerial targets. They can transmit intelligence information from their route. Those drones can also return to base if their weapon systems are not needed. 

"A British MQ-9A Reaper operating over Afghanistan in 2009" (Wikipedia, Unmanned combat aerial vehicle)

The kamikaze drone can also carry things like small air-to-air or anti-radiation missiles. A drone can use those missiles to attack radars and jet fighters. Even a small air-to-air missile can damage the jet fighter. The UCAV can also carry an internal warhead. And that gives them the ability to cause more damage. Internal data links allow the transmission of target information to drones. Remotely. 

Those kamikaze drones can carry other smaller drones, which they can drop into their route. The AI and afterburner will give those drones. An ability to attack aerial objects. Those drones can also operate from any platform, from trucks, potholes, and aircraft. Small drones can be effective against soft targets. Like. Aircraft and helicopters. If. We think about a situation. These small-sized kamikaze drones can slip into the aircraft carriers' hangars. Or drones can also damage aircraft carriers, electronics, and their steam catapults. 

Those weapons can be dropped from other drones or missiles. Or small, remote-controlled speed boats can transport them near those aircraft carriers. Drones can also wait on the beach. They can carry things like missiles, bazookas, and internal explosives. They can close their targets underwater. Rise from the sea. And then slip into the aircraft elevators. Or they can aim their weapons at the radars or the ship’s own missiles. The drone that can drop a bomb into the ship’s missile hatch can detonate the hatch. And then dive itself into that hole, and detonate in the missile tube. 

"The two Global Autonomous Reconnaissance Craft (GARC) assigned to Unmanned Surface Vessel Squadron 3 (USVRON 3) seen here are indicative of the US Navy’s separate ongoing work on smaller USVs. USN" (TWZ.com, Crew Optional Designs Could Be Barred By Law From Navy’s Drone Ship Program)

A drone. Can fly into the air intakes of jet engines. The drone that has image recognition can land from the sky and wait for its target. When a drone. Sees an aircraft or some other target. The AI can give it an order to make a kamikaze mission. The drone can have many types of explosives and warheads. And it can act like mines. A drone can fly through the protective fences and wait near airfields. When it recognizes its target, it can make an attack. The target can be a certain aircraft, tank, other vehicle, or even a person. The image recognition. Doesn’t make a difference between a vehicle and a certain human. 

Drone systems are many times more effective than manned systems. There is no crew; they don’t need things like toilets, water, or food.  Unmanned surface vehicles (USV) and unmanned underwater vehicles (UUV). Those tools that can carry other drones or missiles. Even a small drone can carry anti-tank or other missiles. The USV can have a kamikaze capacity. And it can have a small quadcopter. It can be used for. Search. Targets. 

The small quadcopter that hangs above the USV or unmanned ground vehicle (UGV). The small quadcopter that hovers above them can be wire-guided. That drone can also show targets. For. The drone itself. Or it's laser-guided ammunition. The laser-homing rockets. With. Many types of warheads are installed on jet fighters. But the surface drones can also operate with those rockets. The drone that can point targets to those rockets can make. Precise accuracy. With those rockets. That drone allows operators to shoot those rockets from behind the visual obstacle. And that can make those systems more effective. 

https://www.twz.com/air/american-made-shahed-136-kamikaze-drone-clones-being-tested-by-marines

https://www.twz.com/air/chinas-high-flying-swarm-mothership-drone-has-flown

https://www.twz.com/air/mq-28-ghost-bat-has-fired-an-aim-120-amraam-missile

https://www.twz.com/air/russias-shahed-long-range-drones-are-now-dropping-anti-tank-mines

https://www.twz.com/sea/crew-optional-designs-could-be-barred-by-law-from-navys-drone-ship-program

https://en.wikipedia.org/wiki/Unmanned_aerial_vehicle

https://en.wikipedia.org/wiki/Unmanned_combat_aerial_vehicle

https://en.wikipedia.org/wiki/Unmanned_ground_vehicle

https://en.wikipedia.org/wiki/Poseidon_(unmanned_underwater_vehicle)

https://fi.wikipedia.org/wiki/Tsar-Bomba

https://en.wikipedia.org/wiki/Unmanned_surface_vehicle

https://en.wikipedia.org/wiki/Unmanned_underwater_vehicle

The BCI gives new types of metaverse experiences.

The Neuralink BCI (Brain-computer Interface uses an implanted microchip. That system allows. The brain controls computers using EEG waves. This implant can send to the computer. There is also planned a system where the microchip is in the middle of the brain, just like in the Neuralink system. The system sends electromagnetic waves to the helmet. That covers the head. And then that system can send an echo from that helmet. That makes it possible to create a system that turns brains into a biocomputer. The new BCI systems don’t need implants to communicate with computers. 

Those systems use an EEG that the AI hacked to control computers. The same systems can also control things like prostheses. And also human-shaped robots. This kind of system may already be installed with the new jet fighters. And those systems make it possible. To create synthetic biological brains. Those brains in a vat. It can be created using cloned neurons. Those biological computers can communicate with computers using the BCI. And that makes those systems even more interesting. Than. They used to be. 

The new wireless brain-computer interface systems mean the next-generation communication. Between. Human and computer. This is one of the most interesting and also frightening tools that humans can create. The extremely thin brain implants that can be put on the skull or on the skin don’t necessarily need any advanced brain surgery to bring the BCI to regular users. Even in cases. Where those systems need surgery. Those systems. It might be put on the skull. 

Below the skin, using a regular daytime surgeon. The person who puts those implants must know the point, what the implant must listen to. Another way is to use a helmet. Or. Some kind hat. This helps the user. To put. Those electrodes. Into the right positions. The BCI can turn into a new sense. Today reseachers test the paper-thin chip. That turns light into a new sense for people. 

The BCI system. Transforms EEG into text. And that text can be driven into artificial intelligence. Or algorithms. In those systems, the text-to-speech application was transformed. To use the EEG to create text. The wireless information can be shared through. BlueTooth. The system can use similar communication with an intelligent wristwatch. And that would be the next level. In the human mind. And computer interactions. 

BCI is the tool. That means the human singularity with the machine. The BCI and the AI are tools that allow back-and-forth communication. Straight. With computers and brain shells. This means that the brain will not separate information. That computer transmits into them. From. Real information. And that thing will be the ultimate metaverse experience. This causes a theorem. Maybe in the future, the person. Those who use those BCI systems can get lost in the multiple internal virtual worlds. This brings a new dimension into human life, and that thing is the synthetic digital universe. 

This kind of metaverse is introduced in the Sci-Fi movie “The Matrix”. A metaverse that is created. Using computers is an ultimate platform. It can interconnect people in new ways. And one of those ways is that. The BCI opens the path to the human mind deeper than ever before. The BCI system is one of the things. That can be the next-generation tool for space missions. And the next-generation tool to control human-shaped robots. The metaverse offers a layer between people. Interconnects them together. That layer is a powerful tool. For hackers, a metaverse allows the platform. They can. Use those platforms for their own purposes? 

When we think about the wireless BCI. These kinds of systems. It can be mistakenly connected to the wrong devices. That means hackers can cheat the user. They can. Make a connection with a honeypot, and they can hack that person’s mind.  The BCI is a new tool with strong possibilities. But the threats. The misuse of those systems causes. It is also big. The BCI allows the creation of realistic fake memories. Or the transport of EEG waves. Between people. This thing makes ultimate entertainment possible. But the ability to transmit things. Like remote senses of touch and pleasure. 

Make new ways. To boost the entertainment industry, it can also be used for remote extortion. The ultimate freedom can turn into ultimate slavery. And that is one thing that we must realize. We must realize that laws are not things. That can control the system development. The laws that limit the installation of those components to people other than for medical reasons do not exist in places like Pattaya. There, the boy can get silicone implants in 15 minutes. So what do those doctors do in their back offices?  Some people who are willing to do those surgeries for people who are willing to pay can make them in some places where the police are unable to operate. 

https://www.brown.edu/news/2021-03-31/braingate-wireless

https://www.emergentmind.com/topics/wireless-brain-computer-interactions-bci

https://research.gatech.edu/new-wearable-brain-computer-interface

https://scitechdaily.com/new-paper-thin-brain-implant-could-transform-how-humans-connect-with-ai/

https://scitechdaily.com/scientists-teach-the-brain-to-read-light-as-a-new-sense/

Quantum systems require a new way to think about math.

"IISc physicists discovered that Ramanujan’s classic π-formulas arise naturally in modern theories describing critical phenomena and black holes. The connection suggests his early mathematics may have foreshadowed key structures in today’s high-energy physics. Credit: Stock" (ScitechDaily, Ramanujan’s 100-Year-Old Pi Formula That Hides the Secrets of the Universe)

"A new study reveals that Srinivasa Ramanujan’s century-old formulas for calculating pi unexpectedly emerge within modern theories of critical phenomena, turbulence, and black holes."(ScitechDaily, Ramanujan’s 100-Year-Old Pi Formula That Hides the Secrets of the Universe)

100 years ago, Srinivasa Ramanujan introduced a Pi formula that hides a powerful argument. Pi is a mathematical constant, approximately equal to 3.14159, that is the ratio of a circle's circumference to its diameter. It appears in many formulae across mathematics and physics, and some of these formulae are commonly used. for defining π, to avoid relying on the definition. Of the length of a curve.” Wikipedia, Pi) We know that Pi has a certain value, and we can use that value in mathematical formulae. To use in calculations to determine the area of the circles. 

Pi is also needed. To calculate the volume of things like balls. The ball can be. Determined as a series of circles or rims. So, where is that information needed? What would you do with formulas that can calculate the distance from the ball surface to the layer? That information is useful in the calculations. That is used. For. Calculating the qubit interaction with the receiver. This information is urgent for quantum networking and quantum computing. 

The new quantum systems require new types of calculations. Or those calculations are not. A very new thing. The needed accuracy. The third-degree and higher polynomial functions are very high. The new mathematical formulas must be created. For the quantum systems that require new types of dimensions. And new variables. To make it possible. To simulate quantum systems. For making a complete simulation, the machine requires all information from the system. And the main problem with quantum systems is this. Everything happens in the 3D universe. 

"Choose a point on the circle (blue). You want to map it to a unique point on the straight yellow line. To do this, draw a dashed line between the green point at the top of the circle and your chosen blue point. Then map the blue point to whichever yellow point this dashed line passes through. You can do this for any given point on the circle. (The green point at the top of the circle gets mapped to a special yellow point at infinity.)" (QauntaMagazine, String Theory Inspires a Brilliant, Baffling New Math Proof)

"Unlike in the previous examples, your dashed line sometimes maps two different points on the elliptic curve (blue) to the same point on the yellow line below. You can’t find a map that avoids this, meaning that the elliptic curve has a more complicated set of solutions than the circle or sphere." (QauntaMagazine, String Theory Inspires a Brilliant, Baffling New Math Proof)

A second problem is this. Everything means something in those systems. Things like solar winds, changes in magnetic fields. And other kinds of things. It can have a big effect on systems. There, a superstring travels through a photon. And that turns a photon into a quantum router that shares photonic information into qubits.

Those calculations are needed for models that the systems use for quantum simulations. The millennium problem P=NP (P versus NP) means that checking calculations should be as fast as solving them. Error detection happens. By. Calculating all calculations backward. The problem is in extremely long calculations. If.  The machine uses. Even. Days to calculate some formula. Using backward calculation is a very long process. So if some stage needs more time to check, there should be a problem. But the fact is this. Nobody proved or disproved that problem universally. 

“The P versus NP problem is a major unsolved problem in theoretical computer science. Informally, it asks whether every problem whose solution can be quickly verified can also be quickly solved.”(Wikipedia, P versus NP problem). The idea is that these complex polynomials can be checked by calculating the time. If P=NP. 

Here, "quickly" means an algorithm exists that solves the task and runs in polynomial time (as opposed to, say, exponential time), meaning the task completion time is bounded above by a polynomial function on the size of the input to the algorithm. The general class of questions that some algorithms can answer in polynomial time is "P" or "class P". For some questions, there is no known way to find an answer quickly, but if provided with an answer, it can be verified quickly. The class of questions where an answer can be verified in polynomial time is "NP", standing for "nondeterministic polynomial time” (Wikipedia, P versus NP problem)

The new calculation models give new changes. To solve and check a complex polynomial problem. Things. Like, high-class polynomial functions are new tools. For quantum simulations. New. High-power computers can be used to calculate polynomial functions with new accuracy. This thing requires new accuracy for mathematical constants, like Pi. 

https://www.claymath.org/millennium/p-vs-np/

https://www.quantamagazine.org/string-theory-inspires-a-brilliant-baffling-new-math-proof-20251212/

https://en.wikipedia.org/wiki/P_versus_NP_problem

https://en.wikipedia.org/wiki/Srinivasa_Ramanujan

The brain uses memory Lego to create new behavioral models.

“Princeton researchers found that a primate’s prefrontal cortex reuses modular “cognitive Legos” to solve related tasks, giving biological brains a flexibility that AI still lacks. The insight could help improve AI systems so they retain old skills while learning new ones. Credit: Adapted by Dan Vahaba (Princeton University), from “Brain Silhouette 2” (Littleolred, CC0 1.0, freesvg.org) and “Lego bricks” (Benjamin D. Esham, CC BY-SA 4.0, Wikimedia Commons).” (ScitechDaily, Your Brain Has a Learning Shortcut AI Can’t Copy)

Behavior is the reaction that we see from the outside. And. When an actor notices something, that something acts as a trigger. The trigger launches. A certain behavioral reaction. When we see our friends, we can say “hi”. Or, when we slip on the ice, we put our hands in a certain position, that protects our head from impact with the ground. This means that. When we see something. Hear something, or feel something that causes a reaction. Does something cause a reaction, and what type of reaction is? That depends on the memory blocks connected to that sense. If. There is no memory connection with sense. That doesn’t launch a reaction that we see as behavior. 

The ability to use complete building blocks to create new behavioral models makes human brains more effective than any AI. The AI cannot mimic that ability. Each memory block is like Lego. And. It gives flexibility and effectiveness. To handle memories. And especially the behavior or reflexes. Those are connected with those memories. 

The idea is that there is so-called macro behavior. Behavior is the thing. That. The memory block activates. The modular structure of memories. And actions. What we see from the outside gives humans incredible flexibility. The macro-behavior is like a puzzle, an entirety that includes multiple smaller bits.  Each macro-behavior module can act as a module for larger-scale macro puzzles. 

This means that the human brain handles behavior like a set of modules, which can connect and reconnect with other modules. If we think that. The human brain includes about 100 billion neurons. And each of them has one memory unit. That means there are 100 billion memory units. That gives very high flexibility and morphing ability. For those. Memory structures. The AI cannot mimic those things because that requires the ability to handle so many memory units simultaneously. 

The ability to create new behavioral modules makes human brains effective. The ability to interconnect those modules is unique. One of the reasons. The reason this thing is so unique is that the neuron is not passive. The neuron knows what kind of data it can handle. So it can tell other neurons. That. It's a neuron that processes signals that come from the eye. This means that the neuron that transmits a signal from the retina can ask the routing for data to the neuron that can handle that vision information. 

This saves energy and time because the signal doesn’t disturb other neurons. The transmitter neuron connects a mark. Like a serial number, to the data packet. This information includes data from the transmitter neuron. And if the route that neurons select is wrong. And the receiver neuron cannot handle that information. It can send that information back to the transmitter. Then the routing neurons can ask where the right receiver is. 

https://scitechdaily.com/your-brain-has-a-learning-shortcut-ai-cant-copy/

How intelligent can a computer be?

We must determine “intelligence”. Before. We start to think about that. Somewhere. Intelligence is determined. As the ability to make a decision. By connecting data. That comes from sensors to memories. Then the memory block is connected to some action. Activates the action. That is connected to the memory block. The memory interconnects senses and actions. This is one way to think about intelligence. The next question is this: can a computer be intelligent? Even if that system doesn’t have deep knowledge? And then again. How to determine deep knowledge? 

Deep knowledge or deep learning can be determined. As a series of thoughts, we try to find something deeper in things like chess. Chess can have a role: As a test. That measures strategic thinking or tactical skills. But do those things have something to do with chess itself? When we play chess, we must not think about those things. We must just. Follow the rules. Rules make things acceptable or unacceptable. But then we can say that chess is not the only thing that measures intelligence or strategic abilities. Even the best chess player doesn’t always know how to make chess programs. And even the best programmers might not be good chess players. They might have no chance to practice.

Or maybe. They like to play.

Some other computer games. Intelligence doesn’t mean that the system or actor knows everything. If. We say: the information only. That means intelligence; then a dictionary book is intelligent. Being intelligent means. The ability to apply information to other things and situations. Consciousness means that the creature knows itself. But then, how does a creature show its environment? Is it conscious? Many times, people say that consciousness. Means that the actor. Like an organism defends itself. But then. Things like bacteria also. Defend themselves.  How high a level of consciousness? Do bacteria have? 

When we try to find deep knowledge. Or on some deeper level in some actions. We must ask, does the action have those deeper levels? Or does the action even include some deeper levels that mean something? That means deep knowledge is a chain of thoughts; the thinker can ask. Why does something have some shape? Why do traffic signs have certain shapes? Or colors. But those things. Have no connection with driving. While we drive, we must know how to react to those signs. 

Or, deep thinking?  Do we need deep thinking all the time? The idea is that we can drive cars without knowing anything about things. That happens in the engine. This means we can drive a car by just pressing the pedals and turning the wheels. And then we must follow the rules. We must know traffic signs. Or, rather, we must know how to react to traffic signs. We must not know why the traffic sign is at a certain point. We must know its name. We must know how to react. Do we face situations? We must think. About. The backgrounds of the things that we make. We do many things without even thinking about why we make something. 

If we try. To solve a mathematical formula. We must know how to solve that precise formula. But we must not know all mathematical formulas. We must know. The rules. And how to solve that problem by following mathematical formulas. The deep knowledge and deep learning. Means that we try to find something deeper in actions. That will not need, or that will not have any deeper knowledge.  Are there. Any deeper levels in all things that we do? 

When we try to make a robot that plays chess, the robot must only know how to move buttons, how to hit other buttons, that white starts, and then other rules like promotion and checkmate. Those things make the robot win chess by following the rules. And that makes it possible to win chess. The game can be accepted by chess referees only if the robot follows the rules. 

But when we think about that game, we must ask: what type of other deep knowledge? Or deep understanding? Does that game need? What kinds of deep levels can we find in those games? And if the computer wins that game, are we dumber than the computer? That is one thing that we should ask from the mirror. The computer beats us in computer games. But does it know how to make the beef or salad? Maybe we make better food than a computer. 

https://bigthink.com/books/blaise-aguera-y-arcas/

The AI involves billions of algorithms.

"New research from Johns Hopkins University shows that certain biologically inspired AI architectures can mimic human brain activity even before training on data, challenging long-held assumptions about how AI must learn. Credit: Stock" (ScitechDaily, Johns Hopkins Study Challenges Billion-Dollar AI Models)

New research challenges AI and its learning process. When we try to train extremely large AI’s and their language models, we must understand that this process requires. Lots of work. When we think about things that we make every day, we must realise that. That we learned all the things. That we do. When we try to teach an algorithm, we face one big challenge. The algorithm works like this. The sensor brings data to the memory loop. 

Which compares that data with the database. And if there is a match. That memory loop will activate some action. The problem is that the action that is described to the system is the key. That activates a process. That is connected to the database. And there is always a possibility that the thing. What the computer sees. Doesn’t have 100% match. 

 With the thing. That is described in the database. The single memory loop. That is connected to a certain mission. It is quite easy to train. But when we try to train robots. That can operate. In an open environment, we are in trouble. There must be a memory loop. For every action that the system makes. The thing is that. The memory loop must stop. If the system wants to drive new data into the loop. A quantum computer that operates in multiple states. Makes that system more effective. The system can drive data to different states, and the loop must not stop. 

That makes things like fuzzy logic more effective. Fuzzy logic means a series of precise logical responses. If we compare that thing with the case. When we move our hand, the fuzzy logic means that there is an algorithm for effectively determining. The position of the hand. 

The memory loop or algorithm is the circle of commands that surrounds all the time. If we make a system that goes shopping for us. We must teach every action that the system needs in that everyday journey to the computer. When we think about a shopping trip and compare. It's with a fighter mission, the jet fighter seems more difficult. Because it's not an everyday tool. 

Image 2) Algorithm. 

The thing. What makes the human brain so effective is that it has billions of neurons. A single neuron is not very impressive. It’s a pack of memory and axons, or connections. The human brain uses billions of neurons. At the same time. The central nervous core. Drives data to billions of neurons in the same moment. If neurons don’t find an answer, they call more neurons to the mission. During that process, neurons search for matching memory units within them. And those memory units activate reactions. 

During that process. The area and number of neurons. Those involved actions grow. Until the action is handled in the entire cerebral hemisphere. If both cerebral hemispheres in the cerebrum agree. Or they get the same answer. They send the data to the other nervous system. If those cerebrums disagree, they call the cerebellum. That makes a decision on which one is right. The cerebellum's mission. It is to cut the endless memory loops. 

And when the brains send information to the nervous system. It waits for a report. That the job is done. The report travels to another neuron. That neuron is called a mirror neuron. The purpose of that is to deny. The information collimation in nervous tracks. 

The brain mimic system has two parts. The part that begins the action. And the part. That tells us that the system made the job. In human brains, there are neuron pairs that make those things. The third neuron is needed for cases. The system is stuck in an endless loop. When the first neuron sends a signal, the system does the job. And then it sends that information. Into another neuron, which tells that the job is done. In data centres, those neurons are data processing units. Those units can be multicore processors. In those systems, cores are in pairs, which makes the memory loops travel between them. 

That with things like robot fighters. The robot that goes to the shop. Makes many more actions during missions than a jet fighter. Normally, we don’t think about that thing. Normally, we learn everyday actions during life. But think about the situation. We will drop from space to Earth, and we must learn everything. That we must do. We need somebody. To tell things how to dress, what a means shop, and what vegetables are. Normally, we know those things because we learn those things “automatically”. But what if we grow in an environment? Where are there no vegetables, shops, or money? 

What if we must tell every articulation?  In our body, what must it do? When we want to raise our right foot, we just do that thing. But in the cases of computers, the system must know. What motors must it use? And where it finds those engines. The engine can have a microchip that tells the main computer. That’s the servo that operates the left knee. Then. The operational algorithm tells what the servo engine must do in certain situations. The local system shares responsibility with the main computer. And that makes the main computer’s operations lighter. 

But when we make programs, we must program every action separately. In the second image, you can see the algorithm structure. One algorithm is not very hard. But in complex AI systems. There are billions of those algorithms. There is a missing loop. That means the system should run those algorithms. All the time. The AI is a multilevel construction. The brain mimics systems. Drive data into as many memory loops as it can. In modern AI systems. The system uses other AIs to teach it. 

https://datascience.101workbook.org/05-programming/01-algorithm/01-basics-of-algorithm-structure/#gsc.tab=0

https://scitechdaily.com/johns-hopkins-study-challenges-billion-dollar-ai-models/