Do AI robots have greed, hatred, and delusion?

Greed, hatred, and delusion are specific afflictions of the conscious mind; material form, which is not the conscious mind, has no afflictions of greed, hatred, and delusion. Although a robot is material form and has no conscious mind, and thus in itself has no joy, anger, sorrow, or happiness, nor any afflictions of greed, hatred, and delusion, a robot does not merely possess material properties—it has also been endowed with crude functions resembling those of a conscious mind, capable of performing the crude functions of a conscious mind, namely the function of recognizing text. This kind of recognition of its belongs to the simple, brainless, involuntary-passive, and restricted type. In short, its recognition function is not produced through thinking with wisdom, and therefore it lacks the nature of wisdom; it can be conventionally said to possess the attribute of delusion.

The recognition function of a robot is equivalent to a portion of the functions of human consciousness, and this function is rather simplified: it can only recognize textual information, and it possesses various simple functions such as collecting, organizing, categorizing, summarizing, storing, and outputting information. However, this information comes from other channels and is all pre-existing, old information, not new information developed by the robot itself. What it outputs is always other people's information, always old, already-existing information, without any new information created by itself. Therefore, a robot lacks the nature of wisdom, and it can conventionally be said to possess the attribute of delusion.

No matter how many dharmas a robot can output, it can never act in accordance with what it says; it can speak but cannot do—it is forever merely one who parrots the words of others. Of all the informational content it holds, it has never realized any of it, and it never can realize any of it. Thus an AI robot fundamentally lacks the functions of the manas: it has no capacity to be the master, and no autonomy.

Some Buddhist learners are similar to AI robots: lacking wisdom, they can only use part of the functions of the mind consciousness, and rather crudely at that. Using what the consciousness has learned from elsewhere, they aggregate, process, organize, categorize, and refine it, then arrive at a conclusion, and take this to be their own wisdom—at the same time, arrogance arises. If this were their own wisdom, then AI robots would likewise have wisdom, and wisdom of great breadth, approaching omniscience within the scope of humanity's existing knowledge. Humans are not as broadly knowledgeable as it is; likewise, humans are not as rich in knowledge as "Baidu" (Du Niang). Should robots and "Baidu" then attain buddhahood before all human beings?

Everyone knows the answer to this: it is impossible. Since it is impossible, then for certain people, the lofty discourse of the consciousness, sounding plausible and well-reasoned on every point, is not at all obtained from their own wisdom, still less is it genuine wisdom; their words and actions cannot be consistent—words arrive but actions do not follow. When the manas has no wisdom and has not realized the dharma, one remains an ordinary person (pṛthagjana). Only those whose words and actions are consistent possess the direct-perception wisdom of a sage—wisdom that wells up from within one's own bosom, wisdom that belongs to the individual. Only this wisdom is extremely precious; only with this wisdom can one transform consciousness into wisdom and ultimately accomplish the buddha path. How one should study Buddhism, how one should cultivate and how one should practice—everyone should be clear about this in their own hearts.