De qualibus Intelligentia Artificialis deficit

In brief, we can define two types of synaptic links, both valid for activating any connection between two elements (e.g., element "A" and element "B"), but fundamentally different in terms of their originating cause:
- One type emerges from a direct sensory model (i.e., from our own perceptual experience), meaning that an event involving elements "A" and "B" establishes their connection based on our personal memories.
- The other type is based on an indirect model (i.e., the logical link has been observed or described on behalf of third parties), meaning that someone who experienced that memory transmits it to us.
Both models can lead to the same connection; however, the first directly shapes the reference synaptic associations (to be more technical "neural spikes") whereas the second assumes that the experience has been impersonally "lived" through.
This differentiation might seem trivial or obvious, but it is not for a machine. While (as of current data) a living organism with a mnemonic system capable of processing acquired information filters it through personal deduction based on its sensory and life experiences, an electronic system lacks the second variable—the personal "sensation" provoked by that information during its existence.
This detail thus represents the core that influences the difference between what we can call an "heuristic" thought and another, more socially oriented, of public domain. The more this gap widens, the greater the gap between a subjective discovery—an extremely personal "intuition"—and an induced reasoning. This manifests both in an individual who, on their own initiative, wants to base their mode of thinking on indirect experiences, and in a machine that operates on data imparted by predetermined algorithms.
However, attention must be paid to the fact that while the human mind can operate in both modes, current artificial intelligence programs must (by their nature) process data with a logic that, although broad in sets and variables, remains rigid in its structure from a strictly computational standpoint. The flexibility of "reasoning" is therefore only apparent and influenced by the user's perception, proportionally to their subjective gaps regarding the question posed.
If we apply the "Information Theory" to all this, we could similarly say that a thought, reasoning, or deduction, the more it deviates from common consensus, the greater the informational entropy contained within it. According to this assumption, a personal intuition is definitely more original and creative than one whose extremes have already been extensively defined and documented.
Indeed, in current algorithms, for the reasons outlined above, such creativity is entirely absent, nor is it expected to be simulated in the near future. Implementing experiential programming is primarily limited in primis and without excluding other issues, by the very nature of these computational models-based on statistical-associative comparison which make any programming of subjective awareness highly difficult.
Thus, this is the fundamental difference between an individual who experiences their thoughts through their own qualia, i.e., their personally matured experiences, and a deficient algorithm lacking these emergent properties in its programming.
It should be clear: this does not mean that we will never be able to implement such connections in future programming logic, but we are equally aware that inserting this missing piece will probably require a profound revision of the current technology in use.

In conclusion, consider two elements "A" and "B" for comparison. We could have:
- that "A" is correlated to "B" (A ---> B);
- that "B" is correlated to "A" (A <--- B);
- but also that "A" and "B" are bidirectionally correlated (A <---> B);
Therefore... are these Attitudes, Associations, or Algorithms truly Bidirectional, Best, and Brave?