The bias in AI is a big problem.

The AI's problem is that it doesn't think. That is the thing that we usually repeat and repeat. That ability to get data without analyzing it causes bias. The thing that makes that model possible is that the AI doesn't know what the homepage involves. 

It sees words and their connections to something, but then the system doesn't realize what those words really mean. The AI, or large language model, LLM is programmed to give answers in a certain time. That means the LLM doesn't compare and search connections to every single word in the document. That saves time, but it decreases the data's trustworthiness. 

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

The study put ChatGPT through 18 different bias tests. The results?

AI falls into human decision traps – ChatGPT showed biases like overconfidence or ambiguity aversion, and conjunction fallacy (aka as the “Linda problem”), in nearly half the tests.

AI is great at math, but struggles with judgment calls – It excels at logical and probability-based problems but stumbles when decisions require subjective reasoning.

Bias isn’t going away – Although the newer GPT-4 model is more analytically accurate than its predecessor, it sometimes displayed stronger biases in judgment-based tasks.

The study found that ChatGPT tends to:

Play it safe – AI avoids risk, even when riskier choices might yield better results.

Overestimate itself – ChatGPT assumes it’s more accurate than it really is.

Seek confirmation – AI favors information that supports existing assumptions, rather than challenging them.

Avoid ambiguity – AI prefers alternatives with more certain information and less ambiguity.

(ScitechDaily, More Like Us Than We Realize: ChatGPT Gets Caught Thinking Like a Human)

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

The fact is that at least part of the results that the AI gets from the net are based on the search engine's page ranks. When we think like this: we teach AI we should give feedback on its results. 

If we tell that the AI gives bad answers that helps it to adjust the results that it uses. The problem is always that for giving trustworthy feedback to the AI we should sometimes be experts. Of the thing that we asked. 

The big problem with AI and other copilots is that those things are made for customers. The product must also please the user in some way. And that can cause the bias. When somebody writes about AI, that person should compare it with other LLMs. The most advanced AI "thinks" that it's the best of all. The algorithms tell it how impressive it is. The more advanced AI means that it's better than other AI's. 

When we think about AI and its operators and owners in the world we must realize that all those companies that develop AI are independent companies. That means those AIs don't have access to the competitor's statistics. So, those AIs are not under one dome. And if they should compare themselves with other AIs like "What is the best AI in the world"? 

Those systems cannot get trusted data. Like how many percent of their answers satisfy the users. That kind of data is not collected. The AI will not ask would the user like its answers or if are they good. Another thing is that if the regular user is not the same as some doctoral user. The AI requires specific commands that involve the right terminology. Without the right terminology, the AI is helpless. 

And then AIs cannot compare those percents with other AI's statistics. This makes the AI think. That it's better than it is. The AI can use Wikipedia. When it should use specific sources like CERN homepages. The AI requires that the person who uses it knows the terminology, and then the misuse of some term in the wrong place in a query can turn the AI's text into a grab. AI will not make a difference in the words "hall" and "hall effect". That thing can cause a situation in which the AI starts to talk about echoes or some aircraft hangars when it should start to talk about resistance in electric wires. 

 https://scitechdaily.com/more-like-us-than-we-realize-chatgpt-gets-caught-thinking-like-a-human/

The Chinese company plans to use a maglev (magnetic) launcher for satellites.

Above: artist's vision of the Maglev (Magnetic Levitation) or StarTram launcher. 

The Chinese company plans to challenge Space X's launch dominance. And plans to shoot satellites to orbiters using maglev- or magnetic launch systems. The magnetic launch system means. The system uses similar technology. That is used in rail guns. And the maglev (magnetic levitation) (or, StarTram) launcher is like a giant version of rail guns. The system can use a lightweight tube there is a series of magnetic accelerators. 

The system can stand on the ground using very high pillars. Or things like balloons can raise that tube's end to high enough. In some models is planned to put that magnetic track to the mountain mountain slope. The system can look like an oil tube or maybe the creators of that system will dig it in the slope. And only the open port is visible. 

"Hypothetical StarTram spaceport. The launch tube stretches into the distance to the east on the right (eventually curving up many kilometers away), next to the power plant which charges the SMES (Superconducting Magnetic Energy Storage). RLVs (Reusable Launch Vehicle) return to land on the runway." (Wikipedia, StarTram)

The magnetic cannon can also connect with a centrifugal launcher that gives the satellite or the satellite's aerodynamic capsule a very high punch at the beginning of the flight. Then magnetic track accelerates it to a higher speed. The Spinlaunch company tries to shoot satellites into orbit using spin launchers. Spin launchers are centrifugal launchers. Where a spinning plate gives enough energy and throws the small spacecraft to the orbiter. The spin launcher is easy to connect to the maglev- or magnetic accelerator.

The launch capsule can also have rockets that ignite at the right altitude. That kind of launching system can shoot satellites to the orbiter rapidly. They can be more economical and silent than current rocket-based systems. The Maglev shooter or "super rail gun" can turn projects HARP and Project Babylon into true. Those space guns can be useful tools for civil and military purposes. 

The problem with maglev launchers is their military options. The magnetic version of the famous, but luckily non-finished Project Babylon, the super cannon that cannon engineer Gerald Bull created for Saddam Hussein can turn into a toy in front of that kind of weapon. The high-performance, large-scale maglev cannon can turn the super cannon. Saddam Hussein planned to shoot grenades from Iraq to Israel into toys. The magnetic cannons have much more capacity than the Babylon-type super cannon that shoots traditional explosives to send ammunition into their road or track. 

Those systems can also shoot FOBS (Fractional Orbiter Bombardment Systems) to the orbiter. 

The FOBS satellite can be very small in size. If it carries something like 155 mm. nuclear grenade. Those small-sized weapons can be even more dangerous than the megaton-class FOBS systems. That the Soviets planned in the 1960s. The maglev launchers can shoot small-size hypersonic gliders HGS (Hypersonic Glide Systems) to the orbiter where they can dive back into the atmosphere. A small-size ASAT,  (Anti-Satellite) weapon can also shoot using the maglev launchers. 

https://interestingengineering.com/innovation/spacex-rival-chinas-maglev-launch-pad

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

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

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

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

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

The world's most advanced microchip was published in Taiwan.

"On April 1, 2025, the Taiwanese manufacturer TSMC introduced the world's most advanced microchip: the 2 nanometre (2nm) chip.(ScienceAlert, The 'World's Most Advanced Microchip' Has Been Unveiled)

Mass production is expected for the second half of the year, and TSMC promises it will represent a major step forward in performance and efficiency – potentially reshaping the technological landscape." (ScienceAlert, The 'World's Most Advanced Microchip' Has Been Unveiled)

Taiwanese corporations introduced the most advanced microchip in the world. This is one of the most crucial technological solutions, but it also affects the political field. The most advanced computers make it possible to create new and effective control systems for communication and data handling. 

Even if quantum computers take their place in computing. Binary computers and binary chips play a strong role in field applications like laptops, robots, and military systems control. The quantum computers will be in data centers. And the binary computers run the software that makes field applications run. The user will use the quantum computer remotely through the internet using their own PC- or MAC computer. 

The most advanced microchips are tools that no Western security leader will want to take to the wrong hands. If the Chinese will threaten Taiwan.

That thing can cause a situation. Where that very effective microchip will go to the Chinese hands. 

The binary microchips will control the missiles and fighters because the quantum computers are too large and expensive. 

The question in computing is, who has the fastest,  most effective, and trusted technology? 

The fast microchips can control more complicated algorithms than the slow microchips. The kamikaze hybrid drones that are like cruise missiles. That can carry anti-radar or even dogfight sub-missiles can be the new tools for drone and military technology. 

The cruise-missile-drone combination can be the most effective tool. The cruise missile can drop smaller sub-drones into its path. Those quadcopters can search their targets and home them independently. 

The idea is that the kamikaze drone can also open its path and make sure. They were hit by shooting the anti-radar missiles against radars. That is on their way. Then the main missile can impact the target using its internal warhead.  Those systems require lots of computing to make their missions successful. And fast binary processors are effective in those system's guidance. 

https://www.sciencealert.com/the-worlds-most-advanced-microchip-has-been-unveiled?utm_source=flipboard&utm_content=topic/technology

The future might belong to more than just one AI model.

This is the continuum for the last writing. When we think about AI and its future, it's possible that in the future. There will be multiple internal systems. that require different accesses. The common AIs are tools that every person can use. In those systems are levels or cells that require special accesses. 

The AI would be like the Internet. There is a global data network to which everybody can have access. Then there are technically separated parts called intranets that involve classified- or some data that requires strong authentication. In the same way, AI can form public and limited layers between people and systems their information is hidden from the public.

There can be: 

1) Public AI that searches answers for to regular questions. And helps people, in everyday jobs. That is a similar thing to the Internet. 

2) Intranet AIs that handle information that can endanger people. That information can involve things like confidential business information, and state secrets. And maybe military and intelligence applications. 

The last one can also involve data that is reserved for official use. The robot soldier's control codes can be in that layer. 

The thing is that AI can be many ways more dangerous than nobody thought. The same application that people use to tag their friends on social media can be used to pick the same people from the streets. That system can also recognize vehicles. So, the AI is a dangerous tool in the wrong hands. The AI's misuse is one of the biggest questions on the Internet. Then we can ask what is misuse. 

Is it the AI that controls the weapon systems that protect the country against attackers? Humanoid robots can take almost all human missions. The same system that mows the lawn can make even more complicated surgery operations. The requirement for that is that the AI can have the right dataset. Or the surgeon can operate with remotely. The system records those movements for the next time used. 

The AI is a complicated thing. The same thing that can endanger people can also save their lives. We cannot predict what kinds of needs people gives to the AI. 

The future of the AI is the network of LLM-style programs.

"Leading scientists predict a future where ‘Collective AI’—networks of AI units that learn and share knowledge—will revolutionize fields like cybersecurity, healthcare, and disaster response. Inspired by sci-fi concepts like Star Trek’s Borg but with built-in safeguards, this democratic AI model aims to promote rapid learning and collaboration without centralized control." (ScitechDaily, The Rise of AI: Leading Computer Scientists Predict a Star Trek-Like Future)

"Scientists envision a future of AI units sharing knowledge like a hive-mind, enabling fast, adaptable responses across fields, without the risks of centralized control." (ScitechDaily, The Rise of AI: Leading Computer Scientists Predict a Star Trek-Like Future)

Maybe that is a reality in the future. 

In the future, AI would be like a network of interconnected large-language models, LLMs. The network-based global AI models will be hybrid systems. The system will be a combination of small language models. In user interfaces the spoken language will be more practical. And because the global AI network is everywhere people can make orders from shops saying for example what they want to eat. 

The robot will get receipts from the net and then collect needed stuff from the shop. The robot's hands RFID sensors tell what it took and then it can calculate the bill. The payment can be made using the telephone or the person's biometric recognition allows them to connect those payments to the telephone bill. 

It might ask if a person wants to walk with it or maybe the customer wants to wait at the cafeteria. The robots are parts of the same network and the robot takes images from customer's faces. The waiter robot continues discussions about food asking, is that the first time when a person makes that thing? Or maybe a person wants to borrow one of the humanoid robots that the shop owns to make that food. The humanoid robots are a good combination with AI. 

When the language models are made using multiple small language models that makes development work easier. The developers can make small language models faster than one LLM. 

A system that can look monolithic from the outside can have multiple independently-operating cores. So if that kind of cell-based architecture requires some new things developers must make the new cell to the multicore system. 

The difference between those systems and existing systems is that future LLMs can share data. That means they are more flexible. The LLM can also learn things in different ways than existing LLMs. Today's LLMs cannot change data in their datasets. The modern LLM learns things in an energy-intensive process. That doesn't support the lifetime learning model. 

The future versions of the LLMs can exchange data with each other. They can connect data to the system and disconnect data freely from the system. That means if the AI doesn't use data in a certain time, it erases that data. In the global networked model, the unit that doesn't use some database can send a query to the network. Does some other LLM require that kind of dataset? 

Then, the network transfers that dataset under the control of the LLM or server that requires that data. The system can transfer data in the networks very effectively. The network-based LLM systems are modular. That means researchers can collect them from small-language models. Or SLMs. The network structure helps to keep the system running even if there is some kind of attack. 

The system that core formed of thousands or even billions of independent language models. The AI network can connect or remove those models when needed. The system can use similar language programs, that can learn things from each other. That means the system can recycle the same code in many places. The ability to connect new databases to the entirety. Makes the system more flexible. 

https://scitechdaily.com/the-rise-of-ai-leading-computer-scientists-predict-a-star-trek-like-future/

The new 3D printed hypersonic Mach 5 capable hypersonic vehicles are coming.

"US military could soon get new affordable hypersonic vehicles with Mach 5 speed. (Representational image)"

"The U.S. Army has approved a fresh $3.1 million funding to make hypersonic vehicles faster and more affordable."

"A team at the University of Arizona College of Engineering is exploring the use of multiple metallic alloys and additive manufacturing to enable fabrication of Mach-X – pronounced mock-ex – aerospace technologies as part of a federal governmental push."

"The team, led by Sammy Tin, revealed that the Mach-X vehicles will travel at speeds faster than Mach 5, which is five times the speed of sound and the hypersonic threshold."

"Researchers revealed that the alloys will be joined via 3D printing using compositional grading, in which the concentration of one alloy on the outside of a component gradually reduces to give way to a second alloy layer beneath. These components can be engineered to withstand extreme heat and stress, and also rapidly dissipate heat and minimize localized hot spots."

"The primary goal of the project is to develop a knowledge base for 3D printing protocols that clarifies the costs and tradeoffs associated with materials and processes. This data will move the technology forward and enable producers to create components in nontraditional shapes that are affordable and practical. "

(InterestingEngineering, US students to 3D print hypersonic vehicles with Mach 5 speed for military might)

**********************************************************************************'

Above: Interesting engineering

The U.S. students created a new 3D-printed Mach 5 hypersonic vehicle for the military. 3D printing systems with new metamaterials are flexible tools. If they are big enough, they can even create ships. The 3D printing systems can use any CAD image. And those systems turn those images into physical tools. 

The benefit of 3D printers is their flexibility. The same system can make aircraft and any parts of them. And the 3D-printed hypersonic vehicles are interesting things. They can improve their flexibility to a new level. 3D-printed carbon fiber structures are not a new thing. 

But now, 3D printers are used to make things like rocket engines. And jet plane turbines.

Those systems are new examples of how 3D printers. And advanced AI, along with advanced X-ray technology make it possible to develop new types of safe products. The holograms can make it possible to test how the part fits into some place. 

The 3D printers can create more advanced. And tough products. 

And that's why. Those 3D-printed hypersonic systems are a good point for point in the technology. That revolutionized manufacturing. 

When this kind of technology becomes more common. The scale of products that the 3D printers can make. The fully 3D-printed aircraft are a big advance in aircraft technology. The flexibility of the 3D printer technology allows the system to produce machine parts and ammunition even in operational areas. 

The 3D printers are suitable tools for space systems. The 3D printers can create almost any part that the system requires. In that case, the system must not store the entire spare part storage. Every single piece. That the system takes up space. If there is no use for some spare parts those things are in storage for nothing. 

https://interestingengineering.com/military/us-military-affordable-hypersonic-vehicles?group=test_b

The new plasma thruster uses water as a propellant.

"Florida-based firm Miles Space has demonstrated a water-fueled electric thruster with very low power demands."(Interesting Engineering, Florida startup tests water thruster, runs on just 1.5W for orbital maneuvers)

"The company tested its technology on a European satellite in September 2024. During the flight test, Miles Space’s Poseidon M1.5 thruster produced 37.5 millinewtons of thrust for five minutes at a specific impulse of 4,800 seconds, while drawing power of 1.5 watts." (Interesting Engineering, Florida startup tests water thruster, runs on just 1.5W for orbital maneuvers)

"The thruster fits into a one-unit cubesat and could be used for applications like descent from low-Earth orbit." (Interesting Engineering, Florida startup tests water thruster, runs on just 1.5W for orbital maneuvers)

There are many types of more-or-less practical plasma thrusters. 

Water is a good propellant for rockets. It's cheap, non-toxic, and a common material. Water is not like hydrogen which requires pre-processing. And that makes water easier to handle than hydrogen which must be separated and then turn into a very low temperature. The engine must just expand the water. Basically, the same thruster can use any other liquid from hydrocarbons to hydrogen. 

The system can heat the liquid using an electron, or some other particle beams, electric arcs, lasers, or microwave systems. In some models, the system uses antimatter to boil propellant. The system requires only electricity to create the system that creates the plasma. The system can accelerate plasma by using magnets. 

Plasma thrusters can get their energy from sunlight or from nuclear reactors. One form of so-called solar- or light-sails is the mirror that focuses sunlight into the rocket's engine chamber. There that system can expand propellant. The parabolic mirror can aim sunlight at the carbon fiber structure. Then the system can inject hydrogen into the chamber. There that carbon fiber structure heats the propellant. 

In the most extreme versions, the system uses a laser that can get its energy from the sunlight. The mirror system collects energy and focuses it on the laser element. Then laser beam vaporizes water at hypercritical temperature. There that water turns into plasma. 

When we think about the most exotic versions of that kind of system the engine can use some kind of electrolytic system. The system injects a water ball into the engine chamber. 

The electrolytic system can break water molecules and then the positive and negative electrodes pull those ions and anions into the different directions. When hydrogen ions travel to the anode and oxygen travel to the cathode that makes it possible to create a very exotic ion engine. The oxygen travels to the plate at the front of the chamber. Hydrogen ions can travel out from the engine through the acceleration tube. 

That forms asymmetry in the power. The positive particles travel to the plate where they can from the push. And then another, negative particles travel back from the system from time there they don't create thrust. Those poles in the system can be opposite. Those kinds of ion engines are interesting tools. 

https://interestingengineering.com/innovation/startup-tests-water-fueled-plasma-thruster?group=test_b