The gain of the I side is 3x higher than it should be. But this is somewhat cancelled out by the extra resistance (3.3K instead of 1K) in series to the I-Q balance trimmer pot. The I-Q balance can usually be adjusted as normal. The net result is that there is hardly any impact. This is why it went unnoticed for so long. There would be a slight decrease in dynamic range. Unlikely to be noticeable in most cases. Generally if you have already built the radio with R19 and R25 swapped, and all is working fine, then I would suggest leaving it as it is. That's what I have done (i.e., left it alone) on both my QCX currently operational here (20m and 30m).
Anybody who builds the kit using manual 1.08 will automatically get the resistors the correct way around.?
There will not be any PCB revision to change the silkscreen in any future PCB manufacturing runs. The manual (1.08 and on) has already been amended to fix the issue.