I have not seen the individual study itself, just news reports. I don't have their exact test methodology, but I can't conceive of any scenario where a 70% sensitive test would result in 96% false positives, so there has to be another answer... namely... there are a large number of asymptomatic positives.