Absolutely! I was merely talking about the purely practical/technical issues of letting it talk to a terminal (or anything more complicated like the internet).
In any case, once there's a decent (official) API we can then have ChatGPT talk to itself while giving it access to a shell: Before forwarding one "instance"'s answer to the other, we would pipe it through a parser, analyze it for shell commands, execute them, inject the shell output into the answer, and then use the result as a prompt for the second ChatGPT "instance". And so on.
In any case, once there's a decent (official) API we can then have ChatGPT talk to itself while giving it access to a shell: Before forwarding one "instance"'s answer to the other, we would pipe it through a parser, analyze it for shell commands, execute them, inject the shell output into the answer, and then use the result as a prompt for the second ChatGPT "instance". And so on.