As you said, they’ve loudly assumed it’s possible—so wouldn’t it make sense for them to draw a line in the sand now, before the horse has bolted, to indicate where the “now it’s a national security matter” threshold lies?

Quantum computing is based on a series of scientific breakthroughs and still needs quite a few scientific and technological breakthroughs in several domains before it becomes viable for cryptography (in other fields, we're much closer), in addition to lots of custom hardware.

It's extremely rare (and unpredictable) for a scientist to achieve any kind of breakthrough entirely on their own. They need to exchange ideas with other scientists from all over the world. So you pretty much need your scientists are to do their research largely in public – it _might_ be possible to emulate this if you have a large enough number of scientists on some kind of secret campus, but you'll need to make sure that you're hiring top scientists and you're hurting their ability to both learn and teach the future top scientists you're also going to need and their disappearance from the public track will attract lots of attention.

Add to this the custom hardware, which will quite often come from another country, and it's really hard to keep the big things secret.