While I agree that the early adopters of this tech will be the most successful, I personally think the best thing we can do is expose as many people as possible to it, and most importantly educating them on the right ways to use it.

Just my thoughts, but I can't see any moratorium working the way you explain. While in other realms of science like biology, we can restrict access to certain chemicals, lab equipment, and biological agents to users based on their knowledge and credentials, the most we can do with AI at the moment is the same.

We can let people play with ChatGPT, Dall.e and the others, in a controlled environment before we move to the more advanced features which will come out in the future, regardless if we want them to or not. That way we create the best legislature regarding its usage.