The transistor turns into an artificial neuron.

"Brain Computer CPU Illustration"

"National University of Singapore, NUS scientists have made a standard silicon transistor mimic brain functions, paving the way for efficient, scalable AI hardware using existing chip technology." (ScitechDaily, AI Breakthrough: Scientists Transform Everyday Transistor Into an Artificial Neuron)

"NUS researchers have shown that a single transistor can replicate both neural and synaptic behaviors, marking a significant step toward brain-inspired computing." (ScitechDaily, AI Breakthrough: Scientists Transform Everyday Transistor Into an Artificial Neuron)

The transistor transformed into an artificial neuron. And that makes it possible to develop new types of AI where the operating system and software-based technology cooperate better in the AI-based solutions. The ability to transform transistors into neurons is the new step in hardware-based AI solutions. It's possible. That the transistor can have a memory block. If researchers equip a transistor with RAM (Read Access Memory) that thing turns the transistor into an artificial neuron. 

The transistor that controls the robot's reflexes can also have ROM (Read Only Memories) that it cannot overwrite. This thing paves the way for in-memory computing, IMC, and artificial (morphing) neural networks, ANN.

The transistor is like a neuron and axon, there are two positions one and zero. The difference between these types of solutions and the the systems of yesterday is that the transistors can operate as large groups just like neurons. That makes the system more flexible and energy-friendly than the present system. 

The transistor-based neural systems can select the number of transistors. That is required for each operation. That means the system can make many different operands at the same time. Or some of those artificial neurons can rest if the system doesn't need them. One of the great things that makes the human brain so impressive is that they have so many brain cells. 

The thing that makes the neurons so powerful is that they operate always as groups. Theoretically, the neural structure is quite simple to mimic. The problem is that they create those connections in real life. The neuron has only two axons. Input and output axon. However, in normal neurons, there are many axon connections. 

A neuron needs those connections to aim the signals and neurotransmitters in the right routes. Nanotechnology allows to creation of lots of artificial neurons. That thing makes it possible to create new and flexible computers. 

Some of those axons are needed to prepare the other neurons for incoming messages. There are at least 100 billion physical neurons in human brains. Those neurons can reconnect themselves into subgroups. Those neuron groups form virtual neurons that act as the real, or physical neurons. The ability to connect memories that neurons have into multiple new forms. Is because of those virtual neurons. That thing makes it possible to connect memories into new forms and connect new skills and actions to those memories.