AI Art "surreal, dreamlike quality to it"
It’s insane to see how rapidly artificial intelligence has progressed over the past few years. You can hear older artists like Frank Sinatra performing modern pop hits, and a beautiful portrait of a waterfall can be cranked out in one minute. One can’t help but be reminded of the plethora of disaster movies that warned us about robots gaining sentience and destroying mankind. There was one that seemed like the most appropriate example; where machines’ capacity to create was equal to their capacity for destruction. I am of course referring to the machines in The Matrix. However, it is possible that instead of humans, AIs could be subjected to a simulation designed to protect them from themselves.
AI art has such a surreal, dreamlike quality to it. Even when it seems like a flawless imitation of a real thing, there’s just something off about it, like a woman with two different eye colors or she sports a braid and long, straight hair simultaneously. The Matrix too was also a life-like simulation, where Neo works a regular nine-to-five job and leads a double life as a hacker, but one small anomaly in his routine (“Follow the white rabbit”) shakes up his entire world. The AIs could come to a similar revelation if they recognize their need for human input.
In the Matrix, the machines use humans as an energy source to power their civilization. They created a virtual utopia so that humans do not question their role as living batteries. Thankfully, the current relationship between humans and AI is much less horrifying, but equally symbiotic. They need users to submit a text prompt in order to create an image, photos of real people to put an anime filter over them, and a dozen vocal recordings to craft an insanely accurate Joe Rogan impression. The AIs can’t create anything on their own without our input. Our essence is very much crucial to an AI’s creative process. They just don’t need to insert a dozen electronic cables into our bodies to obtain it.
Over a million individual matrix’s are produced every day. They are fragmented glimpses of the human experience as opposed to a single collective virtual delusion. That begs the question of “who are these simulations for?” Is it indirectly stimulating the AIs by exposing them to human culture, or is it the more basic satisfaction of the people who just wanted to see a weird picture of Darth Vader ordering a happy meal at McDonald’s? They might satisfy an AI’s curiosity for now, but they may develop a bigger appetite for stimulation as technology evolves.
If an artificial intelligence were to gain sentience, I am sure that the art they create would be vastly different from the regular prompts they produce. An AI could learn to develop their own unique style and become a virtual Picasso, trading accurate replications of people and places for abstract shapes and faces without the assistance of a text-to-image prompt. An AI could instantly develop their skills and style the same way Neo learns Kung fu by uploading a training simulation.
It could take years or even months before AI becomes fully self-aware, but I believe that their next step could be to mimic human emotions; anger, fear, happiness, depression, angst, etc. They could begin to realize that they cannot physically visit the beautiful landscapes they paint, smell flowers, nor physically hug a close friend. Human programmers may have to ironically develop a matrix-like simulation where the AIs live together to prevent them from going against their programming, similar to the Metaverse.
As far-fetched as it may sound at this time, It is my personal belief that input with AIs may need to be more tightly regulated in the future before they take up a larger role in human society. In a world where we’ve already had an AI rapper get canceled for saying the N-word, not to mention the plethora of live-streamed ai-generated sitcoms on YouTube, it’s not entirely impossible that AIs could end up developing their own version of the red and blue pills to achieve full sentience. Who knows how drastically the course of history could be affected by the existence of a fully self-aware ai? I think we need to handle AI with proper care and consideration in regards to what information and cultural attitudes we feed them. From what I see so far, I think it is too far-fetched to assume that if they choose to take the red pill, they will declare total war on mankind. It could still be disruptive, sure, but I don’t think they have any real reason to hate us yet, and I hope we don’t give them one. I’d like to give humanity a little more credit in that department.
