On AI Use

I have been thinking a little bit recently, especially in the light of the question I received relating to AI, on my use of it. I tend to use AI a fair bit now, in a way that divides into three main categories:

• Use at work for writing SQL code, mainly.
• Use recreationally for asking stupid questions, or even occasionally useful questions.
• Use recreationally for doing planning etc. and working out how to go about my hobbies etc.

Initially, actually, as AI (and ChatGPT more specifically; I have never touched any other AI) began to be used, I protested it awhile, and did not use it for general concerns. Eventually acquiescing, as I suppose to an extent my protests were juvenile and unrefined, and given the culture at work of people using it, I have begun to use it more often. However, I don't know if it's been a wise... not choice, but happening.

I don't think it benefits me all that much, and I've been thinking quite heavily of stopping its usage altogether. I'm going to start with more rational (or at least, ratiocinated) arguments, and then go into more emotional arguments afterwards.

Rational arguments

Ethical concerns

There are two main ethical concerns: theft, and the environmental issue.

Of these, I don't really know how to feel about the theft. I don't think intellectual property is a good idea and do believe it should be done away with, however still I believe that what OpenAI et al. have done in the mass collection of huge amounts of work from other people is immoral. Regardless, I believe this is a bit of a moot point, as essentially it's a highly abstracted concern, and is far enough removed from reality it doesn't end up mattering all that much.

A more prominent, and concrete, concern, is the environmental one. Training the AI models requires immense quantities of power, which wouldn't be terrible if it were obtained from good sources (solar, wind, etc.) but of course it is not. It causes immense harm to the environment.

Of course this is not unique to AI, not at all. Agriculture and general computing do the same too, driving a car is the same. However, I also think that each of these is bad, and should be avoided to whatever degree possible. I eat (mostly) vegan for a reason, I do not drive for a reason. I think the environmental argument is very strong here.

Psychical health concerns

AI of course is known to cause a degree of stupidity in the user, for instance in the form of cognitive debt. So, if I have concern for my intelligence, I should probably avoid AI.

I note that in the study it also mentions that the brain-using group felt more satisfied, too. They also felt more satisfaction in the outputs of their work. I notice this too: I am fairly proud of a bit of automation I did at work recently that I wrote myself by hand in SQL. I feel far less proud of the SQL that AI has generated for me.

In fact, I've noticed this myself. There are many SQL concepts my grasp of is fairly weak, for example cross applies. I know if I were not to use AI, I would have a stronger grasp of the concept, having to apply them myself.

A more general concern with AI is that it offers a simulacrum of sociality. Talking to AI is like talking to a person, and tickles that part of a person, without offering the genuine benefits of talking to a real person. It falls in the same category as other modern technologic inventions, such as social media (which offers a simulacrum of group social dynamics and community) and pornography (of course not new in the general sense, but new in the modern sense; offering a simulacrum of relations with the opposite sex). The more painful aspects, which are the also the most intense, and the most gratifying to the soul, are thoroughly stripped. For each:

• Social media does not allow the individual to be involved totally with, or subsumed into, the community, which is the most taxing to the individual, but also the most rewarding aspect of a community, and the reason for desiring it.
• Pornography does not provide the totality of the sensuous experience of sex (only the visual and audial are provided; the olfactory, tactile, gustatory are completely absent, as are the social, romantic, etc. factors). In doing so the sexual experience itself provided is lacking, but also the associated pain following from it: childbirth, the difficulty of rearing a child and committing oneself to one's partner and child, are all lacking. Yet again, these are the most intense, and most important, aspects.
• AI may provide something akin to friendship or collegiateness, but it does not contradict you forcefully, it never says the wrong thing at the wrong time, it never offends on the path to bonding, yet also it never bonds, there is nothing to dive into. However these are most satisfying parts of a friendship.

In this way, given it provides such a simulacrum, I think it makes sense to avoid it. I ought pursue the real thing, no?

(I could reference the Monogatari series here, which has dealt with the topic of the honmono real and the nisemono fake: yet, was not Tsukihi Araragi, a fake through and through, real in her presence as real? AI cannot be real as it does not present reality; it presents simulacrum.)

Vague expectation

When I go to ask ChatGPT a question, I wonder what the benefit of receiving the answer is. Firstly, the truthfulness of the answer is debatable, but let's leave that to one side. What do I do with the answer? Often, either I knew the answer anyways, could easily have found it, or needn't have known the answer anyhow (and likely shan't recall it for any period of time).

Would I be better just not asking the question?

It seems to promote a form of scattered approach to knowledge, where sporadic answering of just-popped-into-my-head questions would be where knowledge is obtained, where of course, I would require a form of serious study, or practice, or best of all, praxis, to genuinely maintain a piece of knowledge.

More emotional arguments

I want to be more disciplined, and I can tell a bit of a foul taste forming in mouth about using AI. I think generally, if I don't like it, whether the arguments are more sound and reasoned, or just a pure dislike, given that life is short, and I cannot see any beauty in using it, I should just stop. If I think of it like a vice, well, stop then!