quick deep learning sim...

[mepyyeti]

on github

So this code is the result of an after class 'discussion' with a prof. Can AI (already devolved into 'deep-learning') escape the Turing test constraints (not pass, mind u lol)? I say no, he says yes. 'self'-repairing software as example...I argued that even that is ultimately loop based, not biologically cognitive (as living organisms are).

Anyway, this is my first attempt at 'articulating a point through code' (paraphrasing Satoshi Nakamoto - aka Shinichi Mochizuki!!!) . My wistful hope is others will use this bit of code to debate if AI can ever be organic...I say no (sorry lol).

#!usr/bin/env python3
#useless-dl0.py

def teach():
	i=0
	my_list = []
	while True:
		print(my_list)
		term = str(input('enter content to memorize: '))
		term=term.lower().strip()
		my_list.append(term)
		if my_list[0] == term and my_list[i-1] == term:
			i+=1
			if i == 2:
				print('I\'m learning...slowly, I suspect, but surely.')
			if i >= 3:
				print(f'Success...I remember {term}')
				print(my_list)
				carry_on()
		else:
			print('Oops..best start over...\nOr give up...')
			i=0
			del my_list[:]
			carry_on()

def carry_on():
	while True:
		keep_learning = str(input('Keep learning? [y/n]'))
		if keep_learning.lower().strip() != 'y':
			print('thank you. good bye.')
			quit()
		else:
			teach()

if __name__=='__main__':
	teach()
else:
print(f'no can do. {__name__} won\'t run.') 

[wavic]

I suggest a new discussion with your prof.
Is it possible for a code to examine the environment include the hardware and its instructions and according to the results to rewrite itself to benefit in any possible way from the ecosystem where it runs?