Indeed, so a 2nd test needs to be done, which itself would only be 70% accurate, so you'd have to test again, and again, and again... until you reach a high enough confidence level in the results (hard when it's only 70% accurate)
If you just go with relying on the one-off 70% accurate antigen test to seperate those who have it with those that haven't, then you're playing with a significant number of potential carriers who will be missed.
Until the test gets up to the high 90's% (cf Pregnancy test is 99+%), then for the wider population it's not really attempting to mitigate risk, just risk perception - hence why people are breaking lockdown willy-nilly.
A better test would be for the anti-bodies.. but the data on how long the anti-bodies survive and whether it copes with any virus mutation, is too little at the moment (ironically it's possible that the people who have had it, need to be the ones continually being tested in lockdown (along with essential people), to see how their anti-bodies are doing over time ...).
Anyway.. thats for another thread.