If your own engineers and scientists already have access to almost all knowledge in their field, it seems unlikely that an invention would be so novel that they

If your own engineers and scientists already have access to almost all knowledge in their field, it seems unlikely that an invention would be so novel that they couldn't reverse engineer it. You could have military engineering kept to a more select group, but independent reinvention by researchers or commercial entities should void any secrecy. Otherwise you get silliness like imported Chinese commodity thermal cameras being better than American ones because of ITAR. Or shutting down hobbyist radar projects; if a hobbyist can do it for fun, adversarial military engineers probably can too. The other defense of course is to base our economy on making things. Pour more resources into manufacturing, industrial automation, and standardization so that when a new technology comes along, we can scale it up. An economy based on knowledge restriction is precarious exactly because others can simply ignore our restrictions. We're hamstringing ourselves and it's an illiberal policy.