The Future of Artificial Intelligence: Can machines become sentient (self-aware)

That's my gut instinct as well. It's not the number of transistors that is the problem. It is the transistors themselves. Making an AI will require a whole new technology and a whole new way of looking at computing, IMHO.The first computers were designed to do fast calculations and computer speed has been judged by that standard ever since. We have just made faster and faster calculators. The human brain is not a calculator. The human brain is a master of association, of matching patterns to patterns, voices to voices, etc. The human brain is relatively poor at doing calculations, but it can do pattern recognition better and faster than any computer.

On top of that, these connections can change over time so that some patterns are reinforced while others atrophy. Neurons themselves can change the threshold needed for them to fire. The brain physically adapts to input. If we are going to copy the human brain for AI then we need a system that can physically change over time in response to input.One interesting system is the field-programmable gate array. This processor allows the programmer to change the actual wiring in the chip. You can actually evolve functional arrays through randomly changing the connections and selecting for functions.

quote:

---

Dr. Adrian Thompson has exploited this device, in conjunction with the principles of evolution, to produce a prototype voice-recognition circuit that can distinguish between and respond to spoken commands using only 37 logic gates - a task that would have been considered impossible for any human engineer. He generated random bit strings of 0s and 1s and used them as configurations for the FPGA, selecting the fittest individuals from each generation, reproducing and randomly mutating them, swapping sections of their code and passing them on to another round of selection. His goal was to evolve a device that could at first discriminate between tones of different frequencies (1 and 10 kilohertz), then distinguish between the spoken words "go" and "stop".

​

This aim was achieved within 3000 generations, but the success was even greater than had been anticipated. The evolved system uses far fewer cells than anything a human engineer could have designed, and it does not even need the most critical component of human-built systems - a clock. How does it work? Thompson has no idea, though he has traced the input signal through a complex arrangement of feedback loops within the evolved circuit. In fact, out of the 37 logic gates the final product uses, five of them are not even connected to the rest of the circuit in any way - yet if their power supply is removed, the circuit stops working. It seems that evolution has exploited some subtle electromagnetic effect of these cells to come up with its solution, yet the exact workings of the complex and intricate evolved structure remain a mystery (Davidson 1997).
source

---

This type of set up has always intrigued me. It even seems to have analogy-like features: "In fact, out of the 37 logic gates the final product uses, five of them are not even connected to the rest of the circuit in any way - yet if their power supply is removed, the circuit stops working. It seems that evolution has exploited some subtle electromagnetic effect of these cells to come up with its solution,".

The part that interests me is the ability to rewire the processor on the go. This is something the brain does as well. As far as I know the CPU in your standard home PC does not do this.

You can't let your quad-core AMD Athlon CPU evolve function like you can with a gated processor. In the example I cited earlier there were small circuits that were not connected to the rest of the main circuit. When these small circuits were removed the function ceased. If the small circuits are producing electromagnetic effects as the scientists in the study suspect then I would also suspect that these circuits are somewhat individualized. No two chips will function the same being that there will be small differences in electromagnetic effects with each processor. Now I could be completely wrong here. Biology is my expertise, not electrical engineering. In biology, the hardware is the software. Perhaps this is another feature that future AI will need.I have always viewed computer software as an abomination. Well, maybe not that strong of a word, but something along those lines. Computer programming has always seemed like trying to make rocks fly like birds. The way in which computers work is very different from how brains work, and yet we use programming to force computers to act in a way that our brains can understand. IMVH(and admittedly poorly informed)O, the next great step in computing will be towards computers that need a lot less programming because the architecture is more like us. Just a thought.