Not that I was about to give up but the whole situation stressed to feel a bit annoying, however at the same time I realized that I’m becoming more and more involved in this project and that as more annoying this is as more interesting the whole thing gets.
I did not want to make my next step that obvious so I allowed myself to dream a little. I freed my thoughts and mostly concentrated on my blog fully dedicating myself into finding the truth in this books, injuries, physics and medical reports. And as I was heading deeper and deeper into this magical world the more I was getting the feeling of emptiness… (Luckily not in my world, but in the text that I’m writing.) Even though it seemed entertaining enough to capture reader’s eyes just by its plot I wanted to make the visual part more magical too.
That’s why my next thought was pictures. What if they could randomly change? And as usual from that thought were going a few paths. Changing picture – is great, so can we just search for them in internet save them and then add? Unfortunately, the answer that gave me chatGPT was a strict no. They said something about the copyright things and just refused to help me, but this time I was more loyal to that and let it slide like nothing happened. If I cannot use just google for that I can use special sites where everyone is safe and I would have permission to do that, so I went there. And just for a test drive tried manually as usual just to use this site directly, and to my biggest disappointment it did not work how I wished it would. Pictures there were mostly connected to everyday life so it was just photos, not a single drawing or themed party somehow related to Harry Potter universe. I felt a huge betrayal.
After running through a couple of such sites I realized that I cannot even find something slightly connected to any kind of other universe things. The sad truth was that I could have used just a couple of nice pictures of Neuschwanschtein or spiders(both references from the books) but the atmosphere won’t be right so I gave up on this idea as it was not full filling my expectations too.
It was time for the second option that I mentioned a few paragraphs before. I thought about AI. Of course I use chatGPT as well as google to find the answers for the questions that bothers me, and of course I know about other AIs as Midjourney, thoughts about which were filling my head now. The idea was simple and great – I can generate whatever I want and place it in my blog and it would be perfect. Pictures are generated by accident so there would be never the same picture in my blog. I was very empowered by this thought, but for some reason left it and returned to writing the blog. Maybe I thought that it was the final project so I was not worrying about it too much, but I did what I did and finished the second book.
Finally, I returned to the project. And the game started again. I headed straight to the AIs that would generate me the right picture. This time I was more careful and chose the sites wisely. I looked that they would be free for the users, I searched so that I could have downloaded pictures from it and finally I made sure that the artificial intelligence would generate my pictures in the right style. My dissensions were never that structured and finally I found the one site that I was searching for. It said to generate pictures on the magical theme, it was free and everything that you generate is technically yours, but of course nothing could have gone this smoothly and there appeared a huge but. Site was not providing you with the exact time that you were supposed to wait until it generates a picture. And it is kind of inevitable that free sites would make you wait longer, but for the first picture I waited an hour. I tried a few more just to make the experiment clear and the time was jumping from a few minutes, to hours, to once it just ignored my request or I was too impatient to wait for it.
Maybe if it would be the only problem I would let it slide, even though it would be highly confusing: is something wrong with the program or is it just AI that took so long to generate something. Also the pictures themselves were not to say of a high quality that allowed only second or third picture to be slightly understandable or even relatable to the context of request. That’s why this project was also thrown into the trash bin.