Memristor-based technology can solve AI’s energy problem.

“A new brain-inspired nanoelectronic device offers a glimpse into a future where artificial intelligence hardware consumes far less energy while becoming more adaptable. Credit: Shutterstock” (ScitechDaily, Tiny Brain-Inspired Device Could Solve AI’s Biggest Energy Problem)

The memristor is a resistor. That remember. Its position when electricity is cut. Wikipedia determines memristor like this: 

“A memristor (a portmanteau of memory resistor) is a non-linear two-terminal electrical component relating electric charge and magnetic flux linkage. It was described and named in 1971 by Leon Chua, completing a theoretical quartet of fundamental electrical components which also comprises the resistor, capacitor, and inductor” (Wikipedia, Memristor)

The memristor is the tool. That can answer the AI’s energy problem. The memristor can remember its position. So. That means the position can act as binary memory or even quantum memory in modern computers. Even if one memristor has positions 1 and 0, that means those memristors can act as a hard disk. And in those cases. The memristor group can act as brain cells for the computer. The memristor uses less energy than standard components, and it can be the answer to AI’s energy problem. 

This technology allows the creation of brain-mimicking systems. 

The ability to use memristors as mass memory makes it possible to decrease the need for electricity. These kinds of systems can also help to keep the system’s temperatures low. And it can improve the system’s effectiveness. When we think. About. The use of the DNA-controlled nanomachines as artificial neurons, the memristors in those machines’ manipulators, can revolutionize computer technology. In those axons. The memristor determines if the value in each axon is 1 or 0. In human brains, each axon. Or their ion channels have a value of 0 or 1. 

The thing that makes human brains so effective is this. There are so many neurons. And the second thing is their morphing neural structure. In that morphing neural network, the system can separate a certain number of neurons into different entities. So that means the brain can share missions between neuron groups. 

Those neuron groups, or neuron clusters, can work with different problems. This is why we can walk, talk, and think. At the same time. But when brains see some bigger problems. It can combine those clusters to work. For one purpose. If the system, or its creator, uses  nanomachines as artificial neurons, it can simply create a thing that mimics the brain. 

https://scitechdaily.com/tiny-brain-inspired-device-could-solve-ais-biggest-energy-problem/

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

The AI still cannot think like humans

. 

"An AI model once claimed to replicate human cognitive behavior across a wide range of tasks, sparking excitement about unified theories of the mind. However, new findings suggest its performance may rely more on learned patterns than true understanding, raising deeper questions about how intelligence is defined and measured. Credit: Shutterstock" (ScitechDaily, Did Scientists Overestimate AI’s Ability To Think Like Humans?)

The AI still cannot think like humans. The AI is an ultimate tool to handle large data masses. And.  In these cases, there is no need for so-called deep knowledge. The AI beats humans. When we think about AI. And its ability to solve mathematical problems. We face an interesting thing. The AI handles things better than humans. If we set formulas for it. And determine variables for the AI. But if the AI must find the right formula from the net. And make the calculations, it will not be. So effective at all. Mathematical problems and mathematics are examples of exact sciences. Every single mathematical problem must be solved following the exact rules. Calculation orders are simple and clear. 

But searching. The right formula is not an exact science. The system must compile its orders with the data that it finds. In the cases where the system must answer verbal questions, it’s not very effective. The world is full of mathematical formulas. And that makes it hard to find them. The AI handles things. Like stock market investments. It must only know whether the value of the stock rises or decreases. This means it must handle only two variables. 

But when the system must think deeply. Or, otherwise saying, answer to the question: Why is something done? Or why something happened, the AI has problems. The AI is a good tool in limited areas. When we develop robots that operate automatically in closed areas, those robots don’t require complicated algorithms. The robot can read even QR codes: Where it must turn. Or where is the thing that it needs? There are no surprises in those closed and limited environments. But. When we make programs for self-driving cars, there are many variables that the robot must notice. 

******************************************

When AI starts data processing, it collects data from multiple sources. It takes this first step. But then it doesn’t take the next step. It doesn’t compare the information that those data sources involve. And that makes the AI give answers that have no connection. With topics of the question. 

******************************************

Do you remember everything? What do you need to notice when you drive a car? And imagine what happens if the programmer doesn’t remember to describe things like balls to the autopilot. Things like DNA-controlled nanomachines are more effective. Chemical programming allows them. To make complicated things. But they follow fixed orders. They can learn many things. They can learn to search for things like new types of cancer cells. But that thing doesn’t mean. Those systems can think like humans. For humans, thinking is more than just a reaction to something that activates some database. 

Human thinking involves aspects like feelings. Computers and machines can mimic feelings. They can react to things like tears. But those systems will not feel anything. The thing. That they see or hear activates some kind of reaction. AI can collect data, but it doesn’t process it. Very deeply. When we ask things. Like. Something that is not very common. The AI might give an answer that is completely wrong. It might be about the wrong topics. And that tells us that the AI cannot think. 

It can collect information, but it cannot compare the sources of the information. Or, it doesn’t know the meaning of the information. This means that the system can search a data source. But it cannot confirm that source. The AI detects topics about the homepage. But then. It cannot understand the thing. That. The homepage involves. 

When we think. About the information processing models. We can say that the AI takes the first step. It searches information from multiple sources. But then. It misses the second step. It doesn’t compare those sources. And that makes AI. To give false and ridiculous answers. 

https://scitechdaily.com/did-scientists-overestimate-ais-ability-to-think-like-humans/

DNA-controlled nanomachines are next-generation tools

.

DNA-controlled nanorobots use DNA molecules as programmes. This chemical computer program allows those systems to operate independently. They can. Perform complex operations. Without the need to use a remote control. This makes sense. Those systems. More effective than. Any nanomachine before them ever was. These kinds of systems can destroy tumors. They can open veins. But there is also. A dark side in those machines. 

They can also serve in the military. Nanomachines that find and destroy cancer cells. They can also attack. Against other types of cells. This makes those systems dangerous. In. Wrong hands. Those machines. They can also search. Any kind of cells. And those systems. They can turn into perfect assassins. Those nanomachines. They are invisible to gas detectors. 

And. They can also slip. Out. From. Body. After. Those machines accomplished their mission. That's one way to think about those machines. The fact is that. We decide how to use those systems. Researchers can load antimatter into those nanomachines. And  that makes them a very dangerous tool. 

The same machine that destroys cancer can also serve darkest purposes. Another thing. That we almost ever think. It is this. Those robots can also operate as neurons. They can communicate with microprocessors. This means: DNA-controlled nanomachines. Or. Otherwise, saying: DNA-controlled microchips. They can. Be used. To. Create an artificial brain. That could. Be. More intelligent than humans. 

This is a type of nanotechnology. It can be. A far more. Revolutionizing tool. Then. Nobody imagined. Those systems are powerful tools. In any hands. They can be systems that can create artificial brains for robots. Or they can  form intelligent, self-assembly structures.  Or, they can carry an antimatter bomb into the target. 

https://scitechdaily.com/dna-robots-are-coming-tiny-machines-that-could-transform-medicine-and-technology/

Quantum encryption took a big step. Because of the Talbot effect.

“ Researchers at the University of Warsaw have demonstrated a new approach to quantum key distribution that leverages high-dimensional encoding and a classical optical phenomenon known as the Talbot effect. By exploiting time-bin superpositions of photons, the system can transmit more information while relying on a surprisingly simple experimental setup built from commercially available components. Credit: Shutterstock” (ScitechDaily, Scientists Harness 19th-Century Optics To Advance Quantum Encryption)

Quantum cryptography is a new tool for enhancing the security of communication. In that model, the system connects information to a physical object. It can share information on different routes. And that makes eavesdropping difficult. It can use a certain color. Or a certain image. As. The key that allows the receiving system to access information. 

 But it's also vital for cases where the binary system wants to transform data into a quantum mode. Without quantum cryptography, the system cannot exchange information between binary and quantum states. The thing called. The Talbot effect is the tool. That can make quantum cryptography more effective.  The quantum network can share information to travel on different routes. It can use certain images to encrypt and decrypt information. In a Talbot-effect-based quantum network, it is possible to create quantum superposition and entanglement between quantum dots. And that makes it possible to create a quantum network. But there are also many other ways to benefit from the Talbot effect. 

“Detection of time-bin superpositions with the temporal Talbot carpet. Credit: Maciej Ogrodnik, University of Warsaw” (ScitechDaily, Scientists Harness 19th-Century Optics To Advance Quantum Encryption)

“The Talbot effect is a diffraction effect first observed in 1836 by Henry Fox Talbot. When a plane wave is incident upon a periodic diffraction grating, the image of the grating is repeated at regular distances away from the grating plane. The regular distance. It is called the Talbot length. And the repeated images are called self-images or Talbot images. “ (Wikipedia, Talbot effect)

Furthermore, at half the Talbot length, a self-image also occurs, but phase-shifted by half a period (the physical meaning of this is that it is laterally shifted by half the width of the grating period). At smaller regular fractions of the Talbot length, sub-images can also be observed. At one-quarter of the Talbot length, the self-image is halved in size, and appears with half the period of the grating (thus twice as many images are seen). At one eighth of the Talbot length, the period and size of the images are halved again, and so forth, creating a fractal pattern. Of sub-images with ever-decreasing size, often referred to as a Talbot carpet. Talbot cavities are used for coherent beam combination of laser sets.” (Wikipedia, Talbot effect)

“The optical Talbot effect for monochromatic light, shown as a "Talbot carpet". At the bottom of the figure, the light can be seen diffracting through a grating, and this pattern is reproduced at the top of the picture (one Talbot length away from the grating). At regular fractions of the Talbot length, the sub-images form.(Wikipedia, Talbot effect)

The second image introduces the Talbot-effect, and there could be  millions of possibilities in the encryption key. As we see, the possibilities. It could be the number of quantum dots. The system is used for encryption. Also. Things like a wavelength (color). And the time at which the image remains could be the thing. That helps to create an encryption key. Also. The system can calculate. How many times? In a time unit, the image blinks can be used to create ultra-secure encryption keys. Also, the time between blinks can be a participant. In quantum encryption. The system can also share information between multiple data lines. And then it can collect that information in the points. Of those quantum dots. 

When we talk. About the effectiveness of quantum cryptography, the diversity of methods. It keeps those things safe. If the system uses multiple different ways to encode messages and other data. AI-based intelligent systems can use multiple things. And ways to secure data. In that kind of encryption, the image that the system transmits could be a teddy bear. Then the receiving system sees the dataset that matches the teddy bear image. When the system receives other information that is not delivered in the image form of a teddy bear, it denies that information. This means the image acts as a key that allows the receiving system to open the message. 

https://scitechdaily.com/scientists-harness-19th-century-optics-to-advance-quantum-encryption/

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

AI and customer profiling.

The AI can predict things and act before a person can even open their mouth. This is possible in cases involving individual people. And specific locations, such as certain restaurants. The system must only have a profile of the person, and then it can predict the person's orders. The system can predict the portion. What a person orders. If. It has knowledge of the person’s favorite food. In a limited space and a limited number of actors and choices, the AI can make predictions very easily. In the case that the system knows. If a person hates boiled cabbage, it can make excellent precision. 

If all portions except one. Involves baked cabbage. The system can make a profile. Simply by asking a person about food. The person will not eat under any circumstances. Or the system. It can follow what food the person throws away. But. There is always a possibility that a person will order the portion that involves baked cabbage anyway. If other things on the saucer taste good. 

These kinds of things make the prediction. For a single person. Almost impossible. We can predict. Somehow, how large groups of humans behave. 

We have scientific models. Of. Behavior in certain situations. We can predict how atoms behave in large-scale models. But we still cannot calculate the point of the single atom in the room. 

We can predict how animals behave. We know when birds make their descendants. But the thing that separates us from animals and atoms is that we can suddenly change our minds. This makes this kind of prediction hard. All people. They will not react the same way. In the same things. Things like our experiences and many other things determine how we react. If we sit in a restaurant. We might not buy food for a garbage can. So, if we throw the food. That we paid. Into the garbage, that means we don’t like it. 

For making profiles, the system can follow. What bites the customer puts on the edge of the saucer. This kind of prediction is possible. But. If the AI wants to predict how a person behaves in normal situations, that can be difficult. In cases like stock marketing, the system must only know whether the stock value rises or not, and then it can make a prediction. About how the investor behaves in that area. The idea is to bring profits, and that means if the stock value decreases, the system must not have a large-scale imagination. 

Then, when the stock value decreases, that means the actor sells those stocks. Or they should sell them. If they want financial benefits. This thing requires that the system can handle two variables. 

But then we come to the cases that we call human. Humans act differently from what nobody predicts. We can have the same morning routine every single day for years. We can come to the same bar. Or cafeteria. Every morning for years. The staff might know us, and maybe they know what we will order. But then one day, we will not come to that cafeteria.  We find another place. We might move to another city without warning. And that makes. It's hard to predict. On how a person behaves. In a large-scale environment. There are many variables that the system must notice to predict a person’s behavior. It’s almost impossible in a large-scale environment. 

https://bigthink.com/science-tech/proactive-ai/

U.S. NAVY tested a railgun in the desert.

The problems with railguns are the speed of their ammunition. The ammunition that the railgun sends to trip is so fast that the air has no time to fill the tube behind it. Therefore, the vacuum that forms in the barrel hinders the rapid operation of the railgun. The railgun uses magnetic fields to accelerate ammunition. Most of that railgun ammunition doesn’t require explosives. They use kinetic energy that they transfer to the target. 

The system uses Gauss-Track, where the strongest magnets are at the edge of the barrel. And that accelerates ammunition. This means the ammunition must not be tight. 

Regular cannons require tight ammunition, because the gas from explosives pushes the ammunition forward. If the gas from explosives travels past the ammunition. Its energy will not affect that ammunition. The time that the high-pressure gas is directly proportional to the range of the ammunition. 

The vacuum is the problem with the hypersonic railgun. 

The system could fix the problem. By allowing air to travel in the barrel behind the ammunition. Or the ammunition can have a channel through it. The tunnel through the ammunition helps to fill the vacuum. This means that. There might be. Some kind of ventilation system that fills the vacuum behind the ammunition. The ammunition can also put hover in the magnetic field. Then the magnets will start to pull it. The system will not cause recoil. And. That allows it to keep a very high rate of fire. 

The electromagnetic cannons must not have magnetic ammunition. But. In the case that magnets accelerate the ammunition, there must be something that those magnetic fields touch. There is a possibility that. Sometimes people confuse that system with things like an electric arc. There is also a tested system called a light-gas gun. 

In that system, the cannon pumps things like hydrogen behind the ammunition, and then that system accelerates the ammunition. In that system, the piston presses hydrogen against the plate. that forms temperature. The system presses hydrogen until the plate, and then high-temperature hydrogen is released behind the ammunition. If there is an oxygen that causes detonation, that accelerates the ammunition. The system could be a hybrid of the light-gas gun and the railgun. 

https://www.twz.com/sea/navy-is-firing-its-railgun-again-after-abandoning-it-for-years

https://en.wikipedia.org/wiki/Light-gas_gun

The new production systems can revolutionize drone systems.

The new portable tactical manufacturing platforms fit in the Atlantic containers. Those miniature, portable, AI-controlled drone factories can build 17 to 50+ drones per day. That means that drone factories using 3D printer-based technology can produce custom drones for each mission. The 3D printing technology. It is the thing that will revolutionize warfare. If those systems use plastic bodies, they can use any hard plastic, including garbage, to create those plastic parts. If the system must make only small drones, it could be so small that it fits into a suitcase. 

The AI-based system can make almost everything. If it can find the right data. The drone. Can be manufactured, and the control program can be loaded into those drones from those automated platforms. 

The system simply melts that plastic. The system can use solar power or regular engines.  Or fuel cells to create its energy. The system can use things. Like methane from composters, or nuclear power as an energy source. So it can be connected to portable power plants. Or. It can get its energy from a regular electric network. The system can get drawings for its computer-aided design/computer-aided manufacturing (CAD/CAM) systems from the internet. 

This means that planners. They can sit far away from that physical platform. And. They can be assisted by the AI agents. The system can cooperate with other robots that can collect things. Like metal. And plastic garbage that the system can melt and cast into wires for the 3D systems. 

And in the wrong hands, those systems. They can be more dangerous than any other system before. The container-sized 3D printer factory involves high-temperature printing tools. Those small. Portable. Fully automatic factories. They can create machine parts and even things like assault weapons. The fact is that the fully automatic 3D printer system can be used to make. The custom spare parts for machines. 

"Ukrainian FPV drone with fiber-optic communication channel" (Wikipedia/ Fiber optic drone)

The fiber-optic. FPV drones and humanoid robots can also operate in cases where things like nuclear power plants are damaged. Those systems must not always operate in war zones. But. The same systems. Those who fix the reactor damage can carry weapons. The fact is that. Those systems are required for small modular reactors (SMR) to come into commercial use. Those robots can fix those reactors if their shells are damaged. That denies the radioactive leak. 

But. Those systems can also create Many other things. Like tools. And even rocket engines. Those systems must have laser tools to ensure quality. If those high-temperature tools can be used with things. Like chromium as a source material. The combination of 3D printers, robot arms, and high-accuracy laser machine tools is an effective combination. It’s possible that in the future, man-shaped robots can have 3D printers in their hands. And drones can also operate as 3D printers. So, those systems can make even full-size ships. And the wire-controlled drones. 

They can also fix things like damaged nuclear reactors. The idea is that a regular drone carries a support station that transforms. Optical or radio-based wireless signals to signals that travel to a humanoid robot or drones through optical wire. Those drones. That could be land vehicles. Can operate multiple robots, like a walker, who controls multiple dogs in multiple lead ropes. There is also a possibility. To make a chain. Of the wire-controlled robots. In that case, the robot’s control electronics can be. In the Faraday cage, which protects it against the ionizing radiation. The robot can use the diesel engine as a power source. 

They can be used to make pistons for engines. The CAM system makes tools. Straight from CAD drawings. The system can use holograms to fit those parts. In the right points. 

This means they are useful in many roles in civil and military operations. Container factories can operate in remote areas to support the organizations, rescue teams, and military operations. The advanced computer-aided design/computer-aided manufacturing. (CAD/CAM) Systems are dual-use tools. 

They can be used to make spare parts for chainsaws. Or, they can create drones straight from CAD drawings. They are flexible systems. They can be transported anywhere using trucks, ships, or airplanes. They can be tools that can make more than one product. And the other thing is that. Almost everybody. Can buy this kind of high-temperature 3D printing system. Those systems can operate in backyards. And the data and source materials determine the quality of the products. 

https://www.accessnewswire.com/newsroom/en/aerospace-and-defense/sensofusion-tactical-drone-factory-a-shipping-container-that-builds-50-interc-1147434

https://en.defence-ua.com/weapon_and_tech/wired_fpv_drones_on_optical_fiber_a_dead_end_a_band_aid_or_a_new_technological_breakthrough_opinion-11608.html

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