I don't want to be consumed by AI
In my opinion, the human being is the most versatile animal on the planet: we adapt to changes that would seem impossible to get used to, and we do so very quickly.
But what if there were situations in which precisely this ability makes us vulnerable?
Note: this text contains extreme simplifications. The sequence of events narrated is meant to convey the sense of what I am saying and does not presume to be historically accurate.
Delegation
Existing is complicated.
Nature is, more than a «loving mother», a «fucking whore» (cit. Slavoj Žižek), that tries to kill us at every opportunity it gets (and, by the way, it always succeeds, at least from the point of view of the individual).
For this reason, from the beginning of civilization, we have tried to separate our world from the purely natural one, building an interface in between that would allow us to reduce the amount of suffering necessary for the continuation of existence.
Nobody likes struggling
From the point of view of physical effort, we have gradually delegated actions previously laboriously carried out “by hand” to tools and technologies we built.
Below are some stupid and simplified examples to make clear what I mean.
1 What a great idea those Sumerians had, 7,000 years ago, to build irrigation canals that would bring water to the fields “automatically” instead of carrying buckets of water by hand.
2 What if we took a circular stone, made a hole in the middle, and stuck into it a horizontal axle connected to a container in which to transport heavy objects? Oh look: that is a cart.
3 Let’s make it so that, instead of pushing the cart ourselves, we make it pullable by animals that are physically stronger than us. Or wait, I have another idea: what if we put on it a magical combustion box that moves the wheels by itself? Too convenient.
The result is that we stopped doing things we did not like doing and expanded our possibilities (in terms of available time and capability) into mental and physical activities that we find pleasant to do.
Continuing down this path, we aimed to build bodies that could further free us from arduous tasks such as working in mines or in sewers, but building a body would have proved useless without an artificial brain capable of controlling it. For this reason, we focused on developing this last element.
Thinking is a pleasant effort
One of the activities that has always fallen into the previously mentioned category of « pleasant to do » is thinking.
Until a few years ago, we had delegated to third-party tools only minor functions within mathematical, logical, and artistic reasoning, freeing ourselves from trivial operations and making ourselves free to think, discover, and experiment in more stimulating and creative areas of the unknown.
- Memorize all human knowledge from the beginning of civilization? Impossible: we write.
- Do calculations in our head? Slow, inconvenient, and boring: we use a calculator.
- Carry out repetitive tasks describable through an algorithm? Alienating: we implement them in a computer.
A few years ago, however, something happened that changed this process: while we were trying to give an artificial brain to the slave machines that were supposed to free us from the hardships of the human condition, we realized we had created a technology capable of freeing us not from repetitive and boring work, but from creative work. We created a technology that frees us from what we considered the signature of our “being human.”
We got used to all this in an incredibly short time, precisely as we have always done with big changes (until 3 years ago, it was unthinkable to chat with a magic box capable of conversing like a human being, not to mention the Turing test).
THIS IS A PROBLEM.
To understand why, try to answer these questions: how many of us are still able to hunt in the forest like our ancestors? Who remembers as many poems as those our grandparents learned by heart? Who is able to do mental arithmetic quickly like our parents?
There is, in fact, a small side effect in the process of searching for convenience: everything you delegate, you lose.
It is necessary to carefully select the activities to delegate to the tools we are creating.
It is necessary to choose carefully what we want to stop knowing how to do.
When to use generative AI
I believe this is a question each person should answer based on their own values and priorities. Personally, I consider generative AI a formidable tool for certain tasks, and a condemnation for others.
Using a chatbot to organize your code, debug, get an explanation of the point of a proof you had been trying to understand for an hour, etc., means continuing that search for “relief from the burden of the human condition through artificial tools” already mentioned in the previous sections.
Using the same chatbot as a mental prosthesis to write opinions that are not your own, annihilating your capacity for imagination, or delegating to it the creative and design part while reserving for yourself the mechanical and algorithmic part of the work, is instead the opposite of the constructive process begun by our ancestors. In this case, man becomes the machine, and the machine becomes the creative element.
I would like, as much as possible, not to find myself in the second case. With that in mind, on this blog I will use AI only to translate between English and Italian and as a tool to generate images (and when possible, I also draw the articles’ cover images myself, because doodling is another enjoyable activity that I don’t want to stop being able to do).