That's not a backspace, that's just a nanny filter. I think all bigco LLMs have something similar. e.g. Claude obviously uses this to prevent Fable from doing security work.

It does it quite consistently on higher reasoning.

I believe they have made display answer a tool call so that it can keep updating its answer as it thinks, instead of doing all the reasoning up front, essentially inverting the usual flow. You can use this pattern with any model that supports tool calling.