With ChatGPT and Stable diffusion we have gotten a taste of what AI is capable of doing, not that we necessarily didn’t know this before; We simply experienced it for the first time.
The question to ask at this stage isn’t what ChaptGPT can do, but instead, what isn’t it able to do? It is able to write software, short stories, fictional dialogue, essays, answer any question, is able to edit all of these, seemingly think for itself… The possibilities are endless and industries are imploding, slowly, right now, as we speak.
But AI’s greatest strength, is also one of the its greatest weaknesses. Wikipedia, while few notes of criticisms are hard to come by, there is one at least one that is significant: It’s correctness of it is so high, that hijacking the webpage can easily dispel false information. Additionally, Wikipedia only ever has one particular writing style and expresses the same types of information about many topics. Like GPT, it does not capture the full range of what is possible, even if what is possible to do, may not be factually correct or realistic.
For practical purposes, it is literally unbelievably good at production of text, but being human is not a matter of functionality. We are not here in order to do something. We are in fact here, because of our bio-sociological conditioning and physical makeup and imagining that we are something more, such as a something akin to God, is also just an illusion of our biology, whose base is chemistry, whose base is physics, whose base is quantum mechanics…
GPT works in the same way, but this is irrelevant. Initially, we believed that the universe was holistic, before people like Galileo came along and showed that the universe was not a holistic thing that worked by God’s strength. Then, the belief emerged as a response that our bodies were irreducible, our bodies being now where God is located. This idea was displaced with biochemistry. Then our brains became the source of Godly intervention, before it is now being reduced as well, to smaller components such as the different cortices and neural connections.
What we do not mention however, regardless of our explanations of how it all works, is the essence of what makes us human, which is something that is not present in GPT, and other AIs: Age and Growth. Programming things into machines that make them speak more slowly, or walk with more difficulty, or forget things is not the same thing as forcing them upon the entity in question. These are processes that occur within everyone, perhaps at different time intervals, but they happen. Weakness and limitation, as it turns out is human. Programming a machine for perfection, whatever that may mean, would make the machine inhuman.
Sure, if one programs these things into a computer. But then the programming of the machine would be human, while the machine itself, would still be far off from actually being an autonomous creature. People change over time, can accept new information, reject it, forget it, invent it, deny it… AIs, currently, and likely for a long time, will not be able to do this, unless we implement these things in them, and if we did, they’d still only be artificial. Silicon based things cannot age in this respect and failure to understand a few even very basic concepts, such as sex, hunger, thirst, and even safety, dooms it forever into still being a machine.
We see then that artificial intelligence cannot displace human beings, because even if we had a perfect AI, it could not understand sex, or hunger, but it could only simulate it. This way, it might give the illusion of being human but don’t be fooled: it’s only executing instructions. CPUs and GPUs that cannot function without its other components. We can live without an arm or leg even without parts of our brain and still achieve similar goals of “extremity-ful” people have. But the machine would shut down immediately if any one of its components were to fail, or be chipped off.
Besides, ulterior motives of people can be spotted in the biology of human beings. Machines do not have that limitation, which means that while AIs may speak kindly and reasonably with us, its inner thoughts will always be a total mystery to us. We are human because we cannot lie, because we are weak, because we age. Anything else is inhuman.
Another Useless Blog 