[darubberduckie]

For the "ask" phase, I feel like that the chat interface makes sense as the first step for the adjust phase because ChatGPT is a LLM, which works seamlessly with any natural language prompts.

But it's a bad interface for most applications. It took me out of context on the work I'm doing and forces me to describe what I am working towards in natural language. I will still use it now because it is a big step forward from Googling. But I think the bigger challenge (which brings the much better experience) lies in how we could deeply integrate the iterative creation process with ChatGPT into the app workflow itself.

[darubberduckie]

Also, I wanted to add that I see two big obstacles that are difficult to overcome unless we fundamentally change how the current generative AI works:

1. Generative AI is inherently non-deterministic. We have no guarantee if the final product it generates works 100% based on the provided prompt.

2. Generative AI is inherently out-of-date. There are limited improvements on very specific apps that allows an interface for AI to grab its updated information. But its difficult to imagine us having a new iteration of GPT daily or even monthly