Over the last few years, it has become undeniable that A.I has significantly improved in its ability to generate all forms of media. From the initial Will Smith eating spaghetti video, to the now generated pictures of landscapes that are indiscernible from reality, it becomes a true area of interest when discussing the predicted trajectory of artificial intelligence. The rapidly accelerating pace in which artificial intelligence can self-improve is almost concerning when considering the potential ramifications that could be manifested through this type of machine learning. Less than a few years ago, AI generated media was blatantly obvious to anyone and there happened to be a stream of guides to help others determine whether these different types of media were false or generated. However, times have quickly changed and since the hallmark of artificial intelligence is its ability to “self improve”, generated artwork has improved exponentially, at an almost scary pace.
In its infancy, artificial intelligence more often than not generated anatomically incorrect human beings. It was glaringly obvious flaw that anyone could quickly pick up on and for a short period of time, was a fool proof method to determine generated media online. Unfortunately, the innate nature of artificial intelligence has forever changed the game. Now there are now even videos on YouTube with side-by-side comparisons asking the viewer to determine the genuine and generated media it is showing the viewer. Initially, I was able to quickly tell the difference between reality and generated. However, as time has gone on, every single LLM has improved to the point at which am I no longer able to determine authenticity with generated media.
For instance, here are two images. One I took of a real mountain formation in the South West and the other I generated, thanks to Google Gemini. Which one to you looks more real?
Photo by NNicols. Originally captured in Fredonia, AZ June 12, 2026. Published July 18, 2026 On nnicols.net
Photo by Google Gemini (Nano Banana 2). 2026, July 18. AI-generated depiction of Monument Valley. Generated via Google Gemini.
Lastly, here is a picture of the real Monument Valley taken by W.Bulach and is Creative Commons.
“Monument Valley” by W. Bulach, licensed under CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0). Sourced from Wikimedia Commons.
So after viewing these images, what do you think this mean for Our Future?
For myself, purely mixed feelings. I’ve come to the conclusion that artificial intelligence has effectively become the next step in the industrial revolution of humanity. While the rate of which it improves may not necessarily be on track with what large investors desire, it has been soundly determined that more raw processing power/investment is a must and entire environments are being transformed to provide for this demand. It is uncharted waters and while I can see the good that can be found in these novel abilities, I also wonder if there are other predominantly negative uses are that will be employed with it. I genuinely wonder if we’ll end up in a zero-trust society in the future considering AI is on track to create media that has no visible flaw in comparison to authentic depictions.
Nathan's Blog 