Artificial Intelligence

I have created the visualizations in this website with a Python program I created. However, I am quick to point out that I am not a programmer or an astronomer. I am a civil and environmental engineer by profession. I was also educated in anthropology, and I have been an amateur artist all my life, all of which helps. I started my project by "naively" asking ChatGPT (I think it was GPT 3) in the Fall of 2024 if it could help me create a digital orrery. I was thinking of a digital orrery program I downloaded a long time ago, called Home Planet. It is still a great program and I recommend it. But, I wanted to create my own. To my amazement, ChatGPT told me it could help me do it using Python. I took one course in Fortran in the 1980s in engineering school, but I am no programmer. But, I said, sure, can you help me do that. ChatGPT then guided me in downloading Python, which is a freeware coding language. After a few hiccups I was able to get it working. Then ChapGPT helped me program a first attempt; see the image above. 

Since my first version, I have added functionality over time, first using ChatGPT 3, ChatGPT 3.5, also Gemini, DeepSeek, then Claude 3 and now Claude 4. Now I use Claude 4 almost exclusively for coding and mostly Gemini for general information. The AI assistants have gotten much more powerful even over the last 9 months or so. Problems that I could not solve at the start now I can solve with the new models. I create the improvements iteratively, by asking the AI for coding suggestions, I try them, I improve them. This has been an amazing process. I can do this programming with AI that I could not do before. It is my creation in the sense that I direct the project, I chose what to implement, I make the actual code in Python. However, the AI provides most of the code detail and solves the programming challenges I give it. I simply tell the AI what I want to do in natural English and it converts my request into actual Python code. Many times, the suggestions don't work right or do not do what I want, but this can be solved by iteratively asking the AI for the needed fixes or improvements. 

This process has made me fascinated with AI. The creative experience is amazing and empowering and engrossing. I have listened to some of the leaders in AI speak about AI. One of the clear messages is that the development of AI is going extremely fast. While AI makes mistakes or even misrepresents facts sometimes, what people call "hallucinations" -- people do this too, AI is very powerful and getting more powerful very fast. As it gets more and more powerful, there is a growing concern that it may become unmanageable. This is called the "alignment" problem.