Why does an antique chess console beat Chat GPT in chess?

When we think about those antique ATARI consoles from the 1978 model we always forget that they were not as easy to win as we thought. Those chess programs handled every kind of data as numeric. And the Chat GPT-type artificial intelligence handles that game as visual data. This is one of the things that we must realize when we think about this type of case. Those old chess consoles used very straight, linear tactics. The main difference between modern algorithms and old-fashioned computer programs is that those old programs are linear. And it handles all buttons and movements separately. 

So there is actually a chessbook in those chess programs, that it follows. Those old chess programs were more difficult than some people believe. If you were a first-timer in chess that means you would lose to those consoles. They played very aggressive and straight games against human opponents. The system tested the suitable movements for each button separately button by button. Because the program was linear the movements were made in a certain order. In those chess programs, every movement is determined by the program square by square. The programmer determined the movements for every button and every square separately. And that made those programs quite long. 

Those old-fashioned chess programs have a weakness that if something goes wrong they continue by following that line. There are certain numbers of lines that the program can use. And there is also the end of the line. Those programs can use complete tactics. But their limit is that those programs are fixed. They don’t write their databases and models again if they lose. And that makes those old-fashioned consoles and video games boring. When people learn tactics that it uses they can beat those old-fashioned programs. The limit of those video games is seen in action games. There are always the same points where enemies jump in front of the players. 

Then we can think about things like learning neural networks. Those networks can beat all old chess programs quite fast. The problem is that the neural network must see the game of the console before it can win those systems. AI is like a human. It requires practicing and training. Without knowledge of the opponent’s game, the AI is helpless. There are many ways to teach AI to create tactics against old-fashioned programs. The system can use some modern chess programs and then analyze the opponent’s game to create tactics. 

The other way is the system can analyze the source code and create a virtual machine that it can use to simulate the chess console game. But what do we learn from that case where antique consoles beat the modern AI? Without training the AI is helpless as humans. If the AI has no knowledge of how to play chess, it must search all data including movements of those buttons that make it as helpless as humans. 

Those old-fashioned consoles are RISC applications. They are made for only one purpose. Their code is completely serving the chess game. Modern AI is a complicated system. It can also do many other things than just serving the chess game. And that makes those old consoles somehow difficult to wing, at least when the AI can break its movements and tactics. 

https://en.shiftdelete.net/chatgpt-fails-in-chess/

The gentle singularity: what is the limit of the singularity?

The next step for artificial intelligence is the artificial general intelligence, AGI. That is the tool that connects every computer under one dome. The AGI is the self-learning system that develops its models and interconnects them with sensors that bring new data for the system. That means we can interconnect every single computer in the world in one entirety. We can think that social media is something new. We forget that a long time before Facebooks were the letter clubs. The “post offices” where people can send letters to people, who could be pseudonyms. 

Social media is not a new thing, and Facebook and other applications are the products of a long route that started in Ancient Rome and Greece where wall writing or graffiti was the beginning of social media. Social media interconnects people from around the world. The new thing that the net brought was speed and maybe the price of those systems is low. But as we know there are no free lunches. The thing that doesn’t cost anything can have the highest price. The ability to create singularity between computers brings the ability to share and receive information with new forces. 

And then the new step for AI and computers is the brain-computer interface, BCI. The BCI means the ability to control computers using the brain waves, or EEG. The system can interact with computers and it can operate also between people. This system can interconnect all animals and humans in one entirety. And there are risks and opportunities about that model. If we make things wrong we create a collective mind. There is one opinion. So we interconnect our minds and computers into giant brains. That is a very sad thing. That thing destroys our own creativity.

The biggest problem with social media, AI-based dating applications, and finally singularity is that the system destroys diversity. People want to discuss and date only people who are similar to them. That means our way of thinking starts to turn homogenous. That causes a situation where we have no people who disagree with us. We can hear only ideas and opinions that please us. We take only people who are similar to us, in our social networks. So, in the worst case, we and our networks operate like some algorithm that recycles data through the model. That means we, our team, or our network will not get anything new to our model. We just recycle something if we don’t accept diversity. 

Our mind needs ideas and motivation for making new things. And where can we get those new ideas? We can discuss those things. Or we can get information that some other party made. And then we can work and refine the information that we can get from net pages and other media. Without opponents our productivity and creativity die, because we have nobody who brings new ideas into our minds. 

In some models, the network can develop things by playing games against some other network. The network creates a simulation and then the model tries to fight against that simulation. If a model wins there is no need to develop it. But if the model loses it requires adjustment. And that means the system requires data and then it requires optimism. 

In the novel “Peace on Earth” the author Stanislaw Lem introduced a model where the simulator creates a model and the other fights against that model. The better simulation becomes a model. Until something creates a new, better model. 

There is another way to operate as a network. The network can accept individually operating members. The idea is that every operator that is connected to the network is autonomous. Those subsystems operate autonomously when they collect data. When the network doesn’t need order it can be chaotic. And when an actor sees something that requires a lot of information, the roll call comes over the network. “Everybody stop, the network needs your capacity”. That commands those autonomous subsystems to leave their work and start to solve bigger problems. 

So, the network operates as a whole when it requires that ability. The network can have subsystems and that means as in the case of an extreme crisis those subnetworks create models that should handle that problem. 

Those subsystems can be individual actors. When the individual actors play against each other, that lost actor joins with the winner and starts to develop a model that won. Then the actor couples start to play against each other, and again. The lost team joins the team that won and then starts to develop the tactics that won the game. The actor groups or networks expand when new actors join bigger entities. 

Those subsystems start to play against each other. When some subsystem loses, that means its tactics are lost. Then that lost actor joins the winner's team and gives its capacity to that team, or network. The network always drops lost tactics or action models until there are two networks against each other. And the better wins. This is one way to create the answer and solution for complicated problems. The expanding network could be the thing that brings solutions to many problems. When the network is in chaotic mode actors search data for it. 

You, me, and the language model.

Who has responsibility if people let their thoughts to some AI?

Why do we let AI think for us? The road to this point is long and rocky. When we order the AI to make essays and poems we follow the journey that began a long time ago. When we read essays and poems, made by AI we can say how those things destroy our creativity. At this point, we might say that we can buy a poem book and write that poem from it. 

So, in this case. We simply copy a poem from the book that some great poem master made. And then we can look at the mirror and ask from that image, who made the poem that we just wrote? We wrote a text that some other person invented. So, if we think about this case, and connect AI to that continuum, we see that AI is taking the role of the poem book. In the point of the receiver or reader, it's the same who made that poem. 

Is it some Chat GPT, or is it some Lord Byron? The poem was not made by the person who wrote it on the card. Then we can think about people like Sam Altmann who make more and more advances in AI. We blame them and AI and search for mistakes from them. But then we forget our own responsibility. The user makes the decision to use AI, so we decide whether will we make poems ourselves or will we let some other actors make those things. We have responsibility for things that we make and introduce to people. When we make and introduce some poems ourselves, we face very pithy criticism. 

When we say that people must go to libraries, read books, and do other things, we must be honest. Are we only jealous of people who have tools and skills that we didn't have 20-30 years ago? When we look at work effectiveness we gaze at things like time. 

That a person uses for work. And if the work is done faster, we give that person new work. Would that be an encouraging way to work? If some person does the work faster than others and the work is well done, should we give the rest of the time free for that person? 

Or should we give a new job to that worker? And then order the person back to the office. And take some artificial smile to our faces and then fire that worker because that person makes work better and faster than we do? Or should we cheat that person about poems that this individual worker published on social media? 

We can also remember that person who is part-time working in our company. That means we can use our supreme control and show everybody how jealous we can be. If a person goes to some poem courses at the labor college outside the working time, we can find a new shift for that person. We have some ideal vision of what a henchman should be, and if a henchman does not fit into that thing, we must change that person to fit into that mold. 

That can be crushing. So, it's easier to take the book from the bookshelf. And then make a copy of some of those well-known poems. That means we can say that the person who invented that poem was somebody else. That might be impressive. We didn’t use our own brains for that poem. We made hard work if we took a pen and then copied those words. But it is easier to make the copy using a computer. Or, maybe we find some of those poems from the net and then use copy-paste. Then we must not use our brains at all. AI is the tool that releases our resources from thinking to something else. When we think about cases where somebody makes their own poems, we must realize that every poem makes the first text. 

We decide the easy way. If we want to write some poems or essays, we must sit on our computer. And then we must take the trouble for that text. If we have some other things to do, we have no time to write texts and think about things that we make. Sam Altmann or anybody else than me and you decide if we use AI. That makes our life easier. It leaves our time to have a social life in discos and bowling alleys. But is that the advance that we want? The answer is that the decisions that we make show the road. 

People like Sam Altmann are basically businessmen. They follow the Maslow hierarchy of needs. When our basic needs are filled we want more. AI is the thing that allows us to transfer all our productivity to some computer. And that is the thing that makes AI advance faster than we expected. When the AI satisfies some need, there is another need it must respond to. This is the thing in AI development. AI can make things better than humans. 

Or, we can say that it can make some things better than humans. But then we must realize that AI must also learn new things. There was a story that some antique ATARI computer beat Chat GPT in chess. That thing happened because nobody ever taught the Chat GPT the chess. In the same way, we would lose all chess games against the monkey if we ever played that game. Every skill that AI has is a module. And if the AI has no module for something it's helpless. The AI requires lots of power. The AI, or LLM server requires its own power platform. And when we develop new and more scalable AI systems, we need new and more powerful computers. 

But still, we must realize that the AI that makes everything cannot make things from nothing. Those systems require massive databases and as much power as some cities. That waste heat can also be used for energy production. But the problem is always the temperature. New solutions like biological AI where the microchips communicate with microchips are coming. And in the wrong hands, those systems are dangerous. 

AI can transform anything that we call humanity.

The ability to create babies with customized abilities is the thing that can make more than nobody predicted. Genetic engineering makes it possible to remove things like hereditary diseases from human genomes. Children can have musical skills or some other kinds of things. The system can select gametes from deliveries who have musical skills.  That makes it possible for the system can order hobbies that those people have. There is a possibility that the customized babies have skills that make them beat their parents. 

That thing is not the only thing that the AI can make. The AI can create children with high-level IQ and that is one of the things that the AGI can create. The ability to manipulate DNA makes it possible to order the color of the skin of a human. Basically, researchers can use genetic engineering and artificial viruses to connect things like chlorophyll to skin cells. And the system can control things like the number of mitochondria in the cells. 

But the fact is that the Chinese military is also interested in super soldiers. Those genetically engineered military men would have high IQs, but their loyalty to the central government can be guaranteed. Those military men can have the genome that makes wolves and dogs loyal to their masters. Genetically engineered super soldiers are one thing that causes fear. 

When we think about controlled evolution we face things that when we favor some skills, there are skills that artificial evolution can destroy. Because society favors certain types of skills that means the hobbies that AI or people who use that tool classify as unnecessary. That means those unnecessary skills will not transfer to other people. 

And that is the problem, because, without those transplants, those skills will not be transferred to the next generation. That means that this kind of controlled evolution decreases the diversity of the genetic and hereditary skills. That causes the situation where the only skills that people seem important start to transfer hereditary. There is the possibility that genetic engineering causes segregation. It's possible. Those genetically engineered people start to avoid natural people as companions. And that causes segregation in society. 

Why are we obsessed with AI?

People are obsessed with AI. The question is: why? The answer can stay in our society. We have the attitude that everything must happen fast. That's why we rather use the Internet than books. There are philosophers, home thinkers, etc, who say that we should go to the library and read books. But when we are in working life we have no time to go to the library and then find things. That we need.  If somebody wants people like students to go to the library and read books they must give time for that. 

When we are at work we must be effective. We have no time to go to the library to search for books, and then write some philosophical thoughts about them. People ask why we give our right to think to AI. The answer is that AI makes everything more effective. If we want to be creative that means we are not effective. If we want to become philosophers we must not expect that our society accepts that thing. 

When we think of something alone we are not social and effective. We are alone with our thoughts. And that is not what society expects us to do. Society wants us to make results. When we write something that takes time. And if we use AI we can make much more texts. Quantity replaces quality. Nobody respects the text that we make ourselves using our own words. People respect models that some other person made. 

Those models make it possible to make more texts and the next step is AI. There is no time to make offers by using your own words. The effectiveness means that people use some models. Lots of offers are better than one that a person makes, using their own words. 

When somebody needs information that means information is needed right in the moment. On our working day, we don't even have time to ask the person who sits next to us that person's name. We don't have time to think about things. And another thing that we have is fear. What if we give a wrong answer? 

That is one of the worst fears in modern life. So if we don't have time to think about things, we don't dare to answer. Using our words, and introducing our own ideas. AI is similar to some poem books. We can take a poem book. And then search for some impressive words and copy them to the text. The next step is the use of AI.

We must use things like AI tools. The AI tool is like a secretary that makes our speeches and other official texts. So we can go in front of people and say, here I read a paper that my secretary has written. That offers the escape door to us. If there is a mistake we can blame our secretaries for that thing.  

The same way, if we make referrals about articles and books that we read, those words might be wise. That's true. But those words are not our own words. Maybe Socrates was a very famous and wise man. But that person wrote his own ideas and words. When we make a speech to our ceremonies we should write our own texts. 

I think that people like Socrates and Plato were very intelligent. But if we just loan those texts, and copy them we cannot find new Socrateses. We cannot find a new philosophy. And what we need is the time to think and the time to handle and observe our thoughts. We are so busy that we have no time to go to the library, and read books. If we are wrong we would face blame. We must have time to go to the Gym after work. We must have time to be social. And we must have time to do many things. 

But then we must realize. That we have no time to sit and read. If we want to go to the library to read books we must find time to do that thing. 

If we buy a book or borrow it. But we have no time to read it. 

That book doesn't offer a very big advance in our knowledge. If we want to get information and use it we must open that book or database etc. And then our mind must be ready to receive that data. 

We don't have time to think about things and the consequences of our work. If we don't dare to write things that we think we cannot find new Socrates and other philosophers. If everything that we write and introduce must be so scientifically proven we should realize that those things don't bring advances. 

https://futurism.com/chatgpt-mental-health-crises

What happens when we get AGI?

What does AGI (Artificial General Intelligence) mean? That is the extension of the large language models, LLMs, that can control every data network in the world. Or the system can control physical tools that are connected under their dome. Normal LLM has its domain. The domain is like a state that involves certain actions. Drone control is one domain, and home appliances are one domain. Those domains can have multiple subdomains. The AGI interconnects those domains under one dome or one entirety. So how far are we from that model? 

The answer is more complicated than we can imagine. We can think that the LLM can control things like microwave ovens, but for controlling those tools the LLM requires a socket that it can use to adjust microwave ovens. So the man-shaped robot can use a microwave oven, or the other version is that the home appliances are equipped with a control system that the AI can use to command it. 

When we connect new things under AI control we can face the same thing as when we try to learn to use some new systems. When we buy something new like a microwave oven, we must learn how to use it. In the same way, the AI must learn to use those equipment. And we have two versions for making that thing. 

To use any tool the AI requires a model that it can use in that operation. The model can be in the central server that runs the AI. But where does that server get the model? That is the point. The operator can teach the AI to use the microwave oven as well as the drone. But the system that is connected to the AI can also involve that model. Things like quadcopters must involve programs that control the rotor’s positions. In those cases the operative model is in the robot, or some other thing. The LLM gives orders to robots where they must travel. 

Then the robot can use its internal systems to navigate and move to the location. But orders for autonomic operations are coming from the central systems. This kind of network-based solution is easier for programmers. In those solutions, every single machine that is connected under the LLM domain has its own operational model. The system is modular and each module is independently programmed. 

Basically, if we think that AGI is the tool that just connects multiple devices under one domain, we could do that thing immediately. We can use man-shaped robots that can do almost everything. But the key word is “almost”. 

 Let’s return to the microwave oven. The reason why it’s hard to make that precise thing is the lack of standard user interfaces. The robot must learn to use every single microwave oven independently. That means it must make an independent model for each microwave oven. If there is a system where we can put seconds and minutes separately, the systems where there are only minutes in the timer are not the same. We learn that difference in minutes. But for robots, we must make an independent model of how to adjust the timer. 

Many systems in the world are so easy to use that nobody has wasted time creating standards for them. Easy systems are easy for people, but then we must think about things like the microwave oven. There are button- or toggle timers and that makes them hard to learn. For robots and AI the difficulty is this in the fact all microwave oven models require their independent model of how to use them. 

The robot must connect images from the user manual to the microwave oven’s interface. There is a possibility that if the system does not learn independently the “teacher” or programmer takes an image of the front panel, and then puts the buttons in the right places. Then the AI can learn the rest of the task from the user manual. 

The self-learning networks crack black holes and control drones.

Artist impression of a neural network that connects the observations (left) to the models (right). Credit: EHT Collaboration/Janssen et al. (Phys.org, Self-learning neural network cracks iconic black holes)

Self-learning networks are tools. That can do many things better than humans. The self-learning network has two datasets. That it can be used in that process. The system has models in its databases. And then the tools that can send observations in that system. The self-learning neural network means that the system compares the observation with the model. 

And if that model is different. The system fixes it. The system has the tools that can handle those images as pixels. The system can change those pixels in the model. That can make it fit with observation. And we can use that model with all learning networks. 

The system can create models itself, or it can use humans to make them. Then that system sends things like drones to operate following the model. Successful missions like pizza delivery or some military action mean that the system has a suitable model. But if the mission is not complete that model requires advance. So if something goes wrong the system requires information on what went wrong. And then the operators should fix the model. In the case of pizza delivery, those operators will fly that mission the first time, and then the system creates the model using that data. 

And that is one way to teach the AI and network to deliver pizza to the right place. The system can use the image of a person who ordered the pizza and finds that person also from outside. In the teaching process, the system needs things. Like the minimum flight altitudes. 

Above: SingularityHub, This Robot Swarm Can Flow Like Liquid and Support a Human’s Weight)

Morphing and learning neural networks can make these kinds of drone swarms ultimate tools in medicine, technology, and weapons. The ultimate morphing ability makes those morphing drones the most advanced tools in the R&D works. 

The same software that recognizes vehicles, can recognize people. The person orders pizza at a certain GPS point or point, where the drone can find easily. The person can also give the image to the drone. 

In other cases. The operator can draw the route. To the delivery point on the city map. The drone knows what streets it should follow. 

The drone can scan things like street names. And then it can find places like certain shop entrances. Then drone can start to search for the person who made an order. If the person gives a personal image to a drone it can search that person from the squares. The same system can also find targets for attack drones. The problem with drones is that they are multipurpose tools. And learning networks can make them more fantastic, and more terrifying than nobody believed. 

The Ukrainian strike against the Russian strategic air force tells how dangerous those systems can be. The drone can be installed in things like trucks and the driver must not even know that they are there. The drone can be on the roof of the container. When it enough close to the target it can release that hatch and then those drones can fly to their targets. The thing is that those kinds of systems are far more advanced than in 2020. Those systems are fast to develop and with the morphing and cheap AI, they are absolutely effective. The AI can be hard to make. But it's cheap to use. 

The new drone swarms can operate like liquid. Other drones can transport them to their targets. The drone swarms can act like liquid metal robots in movies. The drone can actually be formed of that kind of robot swarms. So those drone swarms can travel in the form of a drone. 

Then at the target, they can fall to water. And the drone's shell can turn into a liquid metal amoeba. Then that liquid metal amoeba can do its duty. It can close oil leaks. Or remove cancer from the human body. There are lots of applications for those robots that can make the morphing structures possible, 

https://www.bloomberg.com/features/2025-ukraine-drones-explainer/

https://phys.org/news/2025-06-neural-network-iconic-black-holes.html

https://singularityhub.com/2025/02/24/this-robot-swarm-can-flow-like-liquid-and-support-a-humans-weight/