They knew from a theoretical perspective that such an outcome was possible. And of course they were confident of "ultimate" success.

That said, actually manufacturing the device wasn't as much a physics problem as it was an engineering problem. And engineering problems can have bugs in them. If they were 100% positive then there was no need for the Trinity test was there?

At any rate, my comment related again to the fact that doing a "demonstration" for a select group of observers stood a good chance of being ignored. Which also in turn eliminated one of the two remaining bombs.

Which meant using it on a European city and killing hundreds of thousands, if not millions of people. And if that demonstration was ignored, then what? Destroy another city? And another? (Although that wouldn't be possible, would it? We didn't have the material needed to actually build them in quantity.)

This isn't a physics problem. And they didn't "know". Which meant that they would have to gamble.

You want to talk about secret meetings, and if such existed I can almost write the dialog between the politicians and the scientists.

"So you'll meet your deadline?"

"We should be able to."

"Should? And you guarantee that it will work?"

"Well, no, I mean we can't *guarantee* that..."

"So you want me to gamble on you and an untried, untested, theoretical weapon?"

"Well..."