Interesting read. I think you've covered some pretty important key concepts here. The iterative nature of development with AI and the need for human verification. All of the online discourse I've seen about using AI to generate code uses programming languages which don't produce provably correct code. It takes a human to look at it and say, "This is/isn't what I wanted." What you're talking about seems to take that into account. Testing is going to be extremely important in the future if AI starts to gain wider adoption. As I've used technology like ChatGPT, I've found the best way to get what you want is through iterative prompts. Even then, you may not get exactly what you wanted or you may get something totally incorrect. For example, I've seen problems with ThreeJS code generation from ChatGPT.
Metaphysical ontology and epistemology, analog synthesizers, consciousness, techno, Huayan and Madhyamika Prasangika philosophy, and being with friends and family.
Thank you kindly for your comment! And yeah iterative prompts is the way to go. I love how that is possible to do as it actually remembers what happened previously in the conversation, and can build upon it.
I tried many chatbots over the last 10 years and this is the first time when rational iteration-over-time has actually been working. Not perfectly of course, but it’s actually working so well that I’m trusting it more and more these days to give me what I expect.
Really exciting stuff and I can’t wait for it to get even better.
For further actions, you may consider blocking this person and/or reporting abuse
We're a place where coders share, stay up-to-date and grow their careers.
Interesting read. I think you've covered some pretty important key concepts here. The iterative nature of development with AI and the need for human verification. All of the online discourse I've seen about using AI to generate code uses programming languages which don't produce provably correct code. It takes a human to look at it and say, "This is/isn't what I wanted." What you're talking about seems to take that into account. Testing is going to be extremely important in the future if AI starts to gain wider adoption. As I've used technology like ChatGPT, I've found the best way to get what you want is through iterative prompts. Even then, you may not get exactly what you wanted or you may get something totally incorrect. For example, I've seen problems with ThreeJS code generation from ChatGPT.
Thank you kindly for your comment! And yeah iterative prompts is the way to go. I love how that is possible to do as it actually remembers what happened previously in the conversation, and can build upon it.
I tried many chatbots over the last 10 years and this is the first time when rational iteration-over-time has actually been working. Not perfectly of course, but it’s actually working so well that I’m trusting it more and more these days to give me what I expect.
Really exciting stuff and I can’t wait for it to get even better.