Why don't you ask it not to be?

The default style really annoys me with its verbosity and overuse of numbered lists. Describing a tone and length of responses and anything else you want - think step by step, first principals, show don’t tell - all work pretty well.

You can even ask it to only reply with a Boolean true or false - which certainly doesn’t play to its strengths

Actually, that's a by-product of RLHF. A base model is usually not that verbose.

Use custom instructions. I use "Answers should be very concise by default. Never provide disclosures. Answers should get to the point immediately." and get super concise answers. If I want more detail I just follow up with can you expand on X.

Follow up with the prompt "rewrite that so that it doesn't look stupidly verbose" and it will do exactly as requested.

Give this GPT called Briefly a try. Responses come back as bullet point lists.I prefer more verbose, but interesting to see this in contrast.

https://chat.openai.com/g/g-LNsEQH5rz-briefly