AI can replicate itself. And quite soon it might have imagination.

And more things that can turn into reality sooner than we thought. 

The problem with the AI is that the same algorithm that can sort books in libraries can be used to sort the DNA samples. That allows researchers to search for things like hereditary diseases. But the same algorithms can search DNA sequences that make the "Jennifer Aniston" concept neurons able to create abstractions. There is the possibility of making clones of those concept cells that helps us help us think, imagine, and remember episodes from our lives. If those "Jennifer Aniston" neurons are combined with microchips that gives the computer an ability to think abstractions. 

The bio-chips or biological microchips where the computer communicates with cloned neurons can involve those "Jennifer Aniston" cells. That requires that the "Jennifer Aniston" cells are put in the microchip that can keep them alive. Then those cells must communicate with microchips and non-organic computers. 

The speed of human thinking is ten bits per second. Most computers are faster than they are, but human brains are more powerful than any computer.  The reason why the human brains are so powerful is that they are neural networks. When human brains begin some computing operations they start at multiple points in the same moment. 

All neurons are quite close to each other and that means data must travel between them in short distances. The neurotransmitter allows the brain to make the backup copy of the data if it needs the neural tracks for some other more important purpose. When that happens brain will guide the neurotransmitter to another cell, which stores it with a memory block. And when that "more important" case is over brain retakes data from its memories.  

There are billions of neurons in brains. Every neuron has its own memory block. This makes brains more secure. If one cell dies the bite that loses from memory is as small as possible. And the AI mimics human brains and their processes. 

All networks act similar way without depending are they organic or not. So biological networks get and process information similar way as computer-based morphing neural networks. 

The new AI can use the user's computers. This ability makes this tool more powerful because it can create neural networks using its user's computers. The neural networks are computer-sensor combinations whose size has no limits. The AI with the ability to self-replicate itself is the tool that makes it dangerous. The self-replication ability is the thing that gives AI the ability to keep its service even if its central servers are down. 

The AI is mostly an algorithm. Or algorithms, that the large language model, LLM commands. The computer drive is backward and the users don't see anything on the screen. The next step in AI's advance is that it starts to develop itself. The neural networks can collect lots of data from large areas. They can analyze it and then help to develop new microprocessors. 

The AI can collect lots of data from the computers their batteries, and other kinds of stuff. In neural networks, the system can handle multiple things same time. The system can handle itself in its entirety or every workstation can operate separately. 

Neural networks can take on new duties simply by separating one part of themselves and then the sensor can transport data to the system. Neural networks can also read texts from the screen and that means they can use cameras to get data from the net. 

The ability to make a copy of itself gives the AI a tool that can develop itself. The AI can play ping-pong ball with the duty. The other AI uses its replica as the test environment where it tests its changes. The biggest difference is this, when the AI develops other AI it can do that duty 24/7 without brakes. And that can cause nasty surprises. 

https://interestingengineering.com/innovation/worlds-first-living-computer-switzerland

https://www.livescience.com/technology/artificial-intelligence/ai-can-now-replicate-itself-a-milestone-that-has-experts-terrified

https://www.quantamagazine.org/concept-cells-help-your-brain-abstract-information-and-build-memories-20250121/

https://www.quantamagazine.org/new-book-sorting-algorithm-almost-reaches-perfection-20250124/

https://scitechdaily.com/caltech-scientists-discover-the-surprising-speed-limit-of-human-thought-just-10-bits-per-second/

https://www.tomshardware.com/pc-components/cpus/worlds-first-bioprocessor-uses-16-human-brain-organoids-for-a-million-times-less-power-consumption-than-a-digital-chip